Making a Splash: a Heteregeneous Peer-to-peer Learning Object Repository

In this paper, we describe the heterogeneous peer-to-peer infrastructure for sharing learning object. The POOL project builds on the three types of nodes: SPLASH is a freely downloadable application which allows individuals to create metadata and maintain their collection of learning objects, PONDs are bigger repositories of learning objects connected to the peerto-peer network and POOL Centrals increase the speed and breadth of the searches in the peer-to-peer network. The POOL project uses CanCore – a subset of the IMS metadata protocol – to describe learning objects. In the second part of the paper we discuss the future direction of this initiative based on the maturing learning objects community and lessons learned in the deployment of POOL network. We argue that the standardization effort, although very important, currently provides solutions that are too complex. We see the communities where the knowledge is shared to be the main force in the creation of the metadata standards which would support the growth of semantic web. Key to our approach is the development of overlapping ontologies which address global issues for discovery, and local issues for precision classification and usage. We propose a general mechanism which would allow non-expert community members to create their own metadata description in the form of application profiles, automatically updating their tools when changes to the ontological schema occur and still allowing for the exchange of metadata with the users outside of the community using a part of the basic IMS/SCORM/Dublin Core based metadata.