The amount of information that is available on the Web has increased enormously in a very short period of time. This increase of information is desirable, but it has made the problems of the Web as an information space more evident. One solution to these problems is to enhance the way information is encoded so as to enable machines to help human users of the Web to locate and deduce information from this vast information space. This "future" Web has been named the Semantic Web, and has as a core concept to make information machine processable by using metadata or expressing information directly in a machine processable form, with the final goal of making things easier for the human users of the Web. But to create this Semantic Web there is a need to develop a language that can encode information in a way that enables a machine to enhance the processing of it.
This thesis delivers in the first part a formalization of the Web that is needed in order to develop the language for the Semantic Web since it affects the semantics of the language. The second part of the thesis describes the semantics of the language that is to be used on the Semantic Web. It is considered to consist of three main layers, where each of these layers contributes with primitives that are used by higher layers to create a more powerful language. These three layers together provide the functionality and the expressive power that is needed to be able to represent almost any information that is present on the Web today in a machine processable way.
The articles below are snatched from context...sorry!
How to achieve precision on the Web
The difference between the Web and the Semantic Web
This is the full version (but not the final one). Parts 6.1.3, 6.2.6, and 6.3.3 have some loopholes in them.
On the Semantic Web Language (PDF version)