XML Course Dates

10 things you should know about XML

1) XML is a subset of SGML. SGML (Standard Generalised Markup Language) defined a set of specifications for data sharing between large applications, but was extremely complex, and although it was seen as potentially useful, few applications ever really took advantage of it in practise. This led to the joke that SGML actually stood for “Sounds good – maybe later”. However, it did lead to the development of two very widely used systems – XML, for sharing data, and HTML, for building web pages

2) XML is not actually a language, but a “meta-language”. This means that rather than it being a language in its own right, it is rather a tool for allowing companies, industries and other bodies to create their own languages for describing their data. This means that, for example, a company making carpet tiles may create a language to describe its data with fields such as “TileDimension”, “BackingMaterial” and “IsWashable”, and a charitable foundation may have data such as “DonorFirstName”, “DonatedAmount” and “TaxExempt”. As long as these bodies follow the rules for creating their own language, XML will do the job just fine.

3) Because of its similarity to HTML, XML is ideal as a way of moving data across the Internet – for example, it is not typically blocked by firewalls, whereas customised applications using their own data transporting mechanisms might well find this a problem.

4) “XML” is often used as an umbrella term for various related technologies. Other parts of the suite of related technologies include:
     a. Schemas – used for defining the structure of the data you’ll work with (“There must be a First Name, then a Last Name, there may optionally be a Middle Name” and suchlike).
     b. Transformations – used for taking a set of XML-based data and making it into something else, such as a webpage.
     c. XPath – used to navigate through an XML document and locate the particular bits of data needed.

5) You don’t need to pay a fortune for software to edit your XML files. It’s possible to read and edit an XML file in the humble Windows Notepad (although it may not be sensible – XML files can go on a bit!), and a number of companies offer XML editors for next to nothing. For example, you can use the Microsoft XML Notepad for free (http://msdn.microsoft.com/en-gb/xml/default.aspx), or Oxygen XML Editor for $49 (at the time of writing), or XML Cooktop (www.xmlcooktop.com) for free.

6) On many web pages, you will see a symbol like this: This indicates that you can “subscribe” to a feed of the information – news, sports , share prices or whatever. All you need is some kind of feed reader (such as the gadget on the sidebar of Windows Vista) and the latest news or content will wing its way to you automatically. What does this have to do with XML? The content is provided in XML, which is why it’s so easy for it to get from the Internet to your PC, and to be displayed in a way that suits you.

7) XML is in use behind many other technologies – datasets (in-memory databases) in ADO .NET are XML-based, the new standard for storing Microsoft Office files is Open Office XML (note the new “x” in the Office 2007 file extensions, such as “MyLetter.docx” and “MyFinances.xslx) and many applications use XML-based configuration files to store user preferences and options.

8) When creating XML files, there are two types of check that can be made to ensure that it will work and match your needs. The first is a test for the data being “well-formed”. This means that the data matches a set of rules laid down by the World Wide Web Consortium (W3C) about the expected structure of your data. You can think of this as being equivalent to the basic rules of grammar, such as every sentence must begin with a capital letter and end with a full stop – the rules that everyone always follows, come what may… (ahem). The second set of checks ensures that your data matches your own (or your supplier’s, your industry’s or your boss’s) requirements – these are the rules of validation. Data need not be validated – that’s up to you and your needs for your data, but the data must be well formed.

9) There are essentially two ways of “validating” xml (that is, ensuring that the data meets your own requirements) – by using DTDs, and by using Schemas. DTDs (Doctype Definitions) pre-date XML – they were originally developed for validating SGML (see point 1) and have very limited power. They can ensure that the structure of your XML document is as expected, within fairly broad parameters. For example, you can specify that an element (such as ) may be repeated, but cannot specify a maximum number of times it may be repeated. Furthermore, DTDs only allow validation of structure, not data. Therefore, it is possible to say “The document must contain a DateOfBirth element, but not that this element must be a valid date. Schemas, by contrast, are much more powerful. They too allow for validation of structure, but they allow for structures as complex as “Must contain between 1 and 4 phone numbers” and “Must contain either an Account Number or a Customer ID, but if there’s a Customer ID then there must also be a Payment Ref number”. Furthermore, Schemas allow for the validation of content – from “This must be a date” to “This must be a digit between 1 and 100” or “This must be 2 letters, followed by 5 numbers, then a dash, then 2 more letters between A and F”. The downside of all this power and flexibility is that schemas can be very verbose and hard to read.

10) It is a truth universally acknowledged, that an employee who mentions “XML” in conversation at least 7 times per week at work will be promoted within a month…