Vidi: a Lightweight Protocol Between Visualization Systems and Digital Libraries

Achieving interoperability between digital libraries and visualization tools is a difficult problem. To solve this problem, a version of the Open Archives Initiative (OAI) Protocol for Metadata Harvesting called VIDI is proposed. It is a lightweight protocol, which contains only 5 request verbs -- Identify, ListMetadataFormats, ListVisdataFormats, ListTransformers, and RequestResultSet. It is extended from the OAI protocol, which enables its simplicity and wider acceptability. It is flexible, which avoids a rigid architecture in implementation. It is general, so it can apply to all kinds of Visualization Systems and Digital Libraries. But most importantly, it reaches our goal of enabling operability between Visualization Systems and Digital Libraries. The protocol design and implementation details are given. Two prototype systems are implemented to demonstrate the above features. Implementation details are given about ENVISION-ODL and ENVISION-MARIAN. Analysis, evaluation, and conclusions reinforce the discussion of the benefits of VIDI.