As the number of Web 2.0 sites offering tagging facilities for the users' voluntary content annotation increases, so do the efforts to analyze social phenomena resulting from generated tagging and folksonomies. Most of these efforts provide different views for the understanding of various Web activities. Results from various experimental research should be utilized to improve existing approaches underlying tagging data and contribute further to weaving the Web. However, in practice, there are not enough solutions taking advantage of these results. Even though we can mine social relations via tagging data, it proves no worth for users if this data cannot be reused.In this paper we propose a solution for tag data representation which allows data reuse across different tagging systems. To achieve this goal, we analyze current social tagging practices, existing folksonomy usage as well as semantic Web approaches to data annotation and tagging. We survey and compare existing tag ontologies in an attempt to investigate mapping possibilities between different conceptual models. Finally, we present our method for federation among existing ontologies in order to generate re-usable, semantically-linked data that will underly tagging data.