It makes sense now, indeed.
A document in XML has a root and children.. yes.
So if XML is just another representation of data, it does not necessarily mean that data becomes information when poured into the XML format? After all, information is meaningful by itself (and therefore having semantics), as is said here:
XML translates data into a structure and our brain translates that structure into entities that make sense to us.. ultimately conveying to our brain that what is being seen is actually information. I have my doubts on that last part.
When we introduce 1 and 2 into the element <child>, then our brain (like the gestalt theory postulates: http://en.wikipedia.org/wiki/Gestalt_psychology
) translates that <child> as having the 'child' relationship with its 'parent'. We therefore see it as a whole, an entity and not as a 'sum of parts'. Another relationship would be that values 1 and 2 are now associated with the word 'child'. I could say as well <a>1</a>, <a>2</a>. Now our brain knows that 1 is an 'a'. And so forth.
In practical terms however, XML (as is stated in the slides I am reading) represents data like this:
(input)XML file -> (search its structure and represent it)XPath -> (output) XAML, XHTML, XML, CSV. As we can see, another XML itself can be the result of an input XML file. Therefore it confuses me a little bit what the purpose of the language itself actually is. I agree that it is just another representation of data like you stated. In my example, one could say that one takes data, represents it into XML, uses XPath to search in its tree to use its values, to finally represent it into another format.
I will do some more thinking on this.