XML, what for? (1/2) - exploring XML
XML, what for?
This might be a silly question to ask after writing 18 columns on the topic, but there are indeed many more applications of XML beyond what we looked at so far. We mainly focused on the Web and WAP server tasks of turning XML data into HTML and WML documents, for example using XSL transformations. Other applications for XML are:
XML was largely inspired by the great success of its cousin HTML in the area of displaying pages of text documents. Nevertheless HTML is quite limited in its expressiveness for more complex documents: No page breaks can be defined, automatic numbering of headlines and sections is not supported, and any content type beyond plain text needs to be included as an image in GIF, JPEG or more recenty PNG format.
XML with its extensibility allows for mixing different XML vocabularies, using the XML namespace facility, thereby creating the equivalent of compound documents as they exist for many years now in Office software packages from Microsoft, Lotus and Corel. Some of these companies intend to adopt a mix of HTML and XML as an additional or even the native format for their products.
One of the first widely recognized applications of XML were the Channel Definition Format (CDF) and Rich Site Summary (RSS) for describing the contents of a news channel. Not much is known about the channel described, only that its items must be available through the URLs listed in the CDF or RSS file. Providing information about some object that is otherwise rather obscure is a typical property of metadata.
Metadata exists in many different forms. HTML has a head element that can include the document title and meta tags for keywords and a description. All Office documents have a set of properties like author and creation date that gets saved with the content.
XML has the potential to unify all these different approaches to metadata, opening up the possibility to create a gigantic catalog of documents, irrespective of their type and creator application.
Having the metadata external to the native document content would help in indexing documents more efficiently, without the need to transfer the whole document just to get to the small metadata section.
On to databases and applications.
Created: Sep 10, 2000
Revised: Sep 10, 2000