Text-technological application scenarios have become a fixed component in many fields of the Humanities, which produce increasing quantities of digital data in the research process. This trend induces the need for strategies for intelligent management and archiving of digital resources which are not covered by classical content management systems, such as flexible and web-based integration of archived data in retrieval, analysis and interpretation processes. Frameworks with XML based system architectures can offer appropriate support here. Concentrating on the approach of two frameworks, Cocoon (cocoon.apache.org) and FEDORA (www.fedora-commons.org) the lecture discusses the possibilities, limits and experiences of such approaches, based on our long-time research project (gams.uni-graz.at) at the Centre for Information Modelling at the Faculty of Humanities of the University Graz.