Author(s)

S. Psaroulis, C. Vassiliou, D. Stamoulis & D. Martakos

Abstract

Web sites are using information structured in XML trees. XML is currently the most powerful means of expressing data structures in a user-defined manner, since there are no predefined types for the described data. For example, one can describe a vehicle by means of an XML tree like this: Ford Motors 1968 1800 110 The tags that precede the actual data are not part of any kind of dictionary or catalog defined by XML; hence its flexibility allows the user to create meaningful and readily readable information architectures. However, XML’s main advantage is also at the same time its main drawback as far as interoperability in terms of information exchange is concerned. When information from an XML tree has to feed another XML tree, there is no obvious or automatic way for this exchange. The user has to explicitly define how the two different XML-tree data structures can be correlated. Either a translation table should be provided, or the user links the tags between the two XML trees. The latter is what Microsoft’s Biztalk server supports: the user drags- and-drops the tags of the one XML tree to the other one. Despite the visualization

Keywords