Surely XML has been (one of) the hype(s) for a while now, and I’m sure most of us have heard of it by now.  Every company, that wants to be taken seriously, has re-branded its products in such a way to at least suggest XML usage – a bit like the “Internet Enabled” hype a few years ago!  But why all this?

In this article we’re going to have a look at the whole XML phenomenon – the data exchange factor, well formed XML, validating and parsing an XML document, DOM, SAX and a few others acronyms that have started floating around since the “birth” of XML. 

Brief History of Data Exchange

The articulate speaking was a great step forward for humanity – due to this, we humans are now able to understand each other. But do we? Have you ever tried speaking Finnish to a French person? Or have you tried to speak Russian in Tahiti and be understood?  You might have a chance – according to the Gaussian distribution of probability, but that is in theory! :)

Every single computer program processes data – it’s a well-known fact (there may be an exception: Adobe – they seem to only process “e-paper” nowadays :).

Back in the old early days of DOS, when WordPerfect was THE word processor, and Lotus-1-2-3 was THE spreadsheet application, everything was hunky-dory. However, once companies had started realising the potential of personal computers, desktop publishing and so on, and adding to this the Windows 3.x factor, in no time we were overwhelmed with all sorts of “suites” that promised to do everything in one package.  In fact the reality, however, was far from what the label read.  

But hey, at the time Adobe was (and still is!) the name in graphics processing, WordPerfect (after holding its name for quite a while) was still there, Microsoft Office was fighting its way to the top, so we could still use the files saved in previous versions of these products.

The problem, however, appeared when we wanted to “import” an image from Adobe into WordPerfect, or a WordPerfect document into Microsoft Word. Surely, Windows gave us OLE, but in order to be able to embed a WordPerfect document in Word, you had to have both of them installed, which defeated the purpose of “importing”.

Vendors hurried to overcome this by providing a large variety of converters, but as more and more software was being thrown on the market, soon enough it was impossible to have a converter from everything into everything!  Sure, there were a few standards emerging out, (RTF, HTML and PostScript most notably) which in most cases could have been used as an intermediary format, but that didn’t always work!

Imagine a typical company setup – you have the admin system keeping track of the internals, the HR systems to handle employee files, the accountancy system, the purchasing system, the CRM system and so on. If at any time you had to transfer data from one side to another – unless you were lucky enough to have a “hacker” around to manage to make two of these systems “talk” to each other, re-entering the data in both systems was normally the common solution – there was no easy way for data exported by one system to be imported back into another!

Again, a few solutions have emerged (most notably, at that time, the CSV format), and again, most of the solutions were based on an intermediary format. The list could go on…

The problem was there was no definite intermediary format – some word processors could produce RTF but not HTML, other word processors could import HTML but not RTF; some spreadsheets would import CSV data, while some others would expect a fixed-length field format! So really, once you launched your product into the “real world”, there was no way you could be sure that you had covered all the possibilities.