Web Services based on PROLOG and XML

Abstract. This paper describes how the deductive power of PROLOG can bemade available across the Internet using standardized Web service technologies. This facilitates the use of PROLOG as a component of distributed information sys-tems and in many new application scenarios. Since a lot of information available on the Internet is nowadays XML based and since Web service technologies useX ML based encodings, it is both necessary and useful to be able to process XMLdocuments in P ROLOG itself. To make this possible, a new package for SWI-P ROLOG called X2P is introduced, making available to PROLOG many of theX ML processing facilities of the Libxml2 library, which is a very up-to-dateand efficient implementation of most of the current X ML related standards. 1 Motivation and Overview The World Wide Web as we know it today, was initially designed as a platform forinformation sharing. The core Web technologies, i.e. H TTP, HTML, Web servers andWeb browsers, enable the exchange of information in the form of documents. Web browsers soon became standard tools and it was realized that they can also beused as universal clients to information systems, if these information systems expose their user interface using HTML documents. Most modern information systems use sucha Web based architecture. Tools like application servers simplify the development of Web based information systems at the server side.The development and deployment of distributed information systems, which integrate applications that are distributed on an Intranet or even on the Internet, is still a lotmore difficult. The need for such an (enterprise) application integration over the Web was initially generated mainly by business to business (B2B) applications. However,there are a lot of interesting application scenarios aside from B2B.While the traditional Web is concerned with the interaction between application