Information Retrieval techniques vastly differ from retrieval in pure data management systems. But until very recently P2P techniques focussed on supporting simple database-style retrieval functionality only. Since the content offered in P2P networks currently evolves from simple files that can be adequately described by some meta-data tags, to collections of complex documents that need somewhat more detailed descriptions, the development of refined retrieval techniques has to be fostered.

The aim of this working group is to

• identify basic differences and characteristics of metadata-driven vs. content-driven retrieval in P2P networks
• discuss the practical usefulness (e.g. in terms of effectiveness, scalability, etc.) of current P2P techniques for indexing, routing, etc. in the light of sophisticated content-based searches
• spotlight today's most important research questions and novel techniques that might provide viable solutions

Having done this, we will work to sketch a research agenda for the field af P2P Information Retrieval.