About This Site

Sure, my life is not all that exciting and interesting to a world-wide audience that a personal homepage would be absolutely necessary. In fact, the existence of these few HTML pages does have other reasons than my urge to exhibit myself: It's all about technical feasibility. About concepts like XML, XLink, XSL/XSLT, XHTML, external linkbases and so on.

The challenge was to fulfil the following requirements:

• The contents of the entire Website are to be stored as structured data in XML-documents without regard to the optical presentation. This ensures the strict separation of content from form. It also allows automated processing of and access to individual components of the site, which lays the grounds for developing a simple content management system.
• The structure of the whole site should be reflected in one central document (XML, too), which is stored separately from the XML documents that contain the page contents. This should make it possible to change the hierarchy of the site at any time without having to make manual changes in the navigational part of every single page.
• All the links to internal and external resources are stored separately in a single document ("link base"). Advantage: The link base is the only point where changed URIs need to be updated. All other document will reference URIs through the link base. In addition, the use of a central link base makes it easier to automatically generate a collection of links as presented on the page Links.
• The administration of the site and creating XHTML-documents from XML-based contents following standard layout rules must possible with free tools on any platform.

In fact, with XSL(T) and XLink the world of XML offers everything necessary to fulfil the requirements above. Content can be written using any text editor. XSL(T) processors are freely available for the more common operating systems (even for Windows) - if not, a Java-application will do the task.

In this particular case, each XHTML page on this site has a corresponding XML file on my hard disk. An additionally XML file ("SITE.XML") describes the site hierarchy and is used to store all links. (SITE.XML , in fact, is nothing but an external link base according to the XLink-specification; the handling is so elegantly simple that I can hardly understand why this W3C-recommendation has never really gained wide acceptance.)

With an XSL(T) stylesheet, these data sources are finally converted to XHTML pages. The job used to be done using Microsoft XML Core Services (a free Windows component) and a normal batch file. Today, XSLT C library for Gnome does the job. All you need for a free content management system.

(All I would need now is genuine content in order to justify the work I put in the whole project. And, as you might have noticed in spite of all the pictures and colourful frames: Content is what's still missing.)