81.   Data exchange formats: experience from the National Biodiversity Network

Steve B; Wilkinson1, Lawrence Way1, Andy Brewer1, Richard Ostler2 & Jon Cooper2

1 JNCC Support Company, City Road, Peterborough, PE1 1JY, UK

2 Centre of Ecology and Hydrology, Monks Wood, Abbots Ripton, Huntingdon, Cambridgeshire, PE28 2LS

The National Biodiversity Network (NBN) is an initiative that attempts to improve the flow of biodiversity data within the UK between the data providers and the users. The situation in the UK is relatively unique in that the vast majority of the data is actually collected by volunteer naturalists, not publicly funded collection, and often the workers are very skilled at fieldwork but generally do not have equivalent technical skills. One of the key problems identified across the network has been data exchange, both between the naturalists themselves and between them and other users of the data. The NBN have developed a range of solutions to cope with this:

·         Broad, heavily attributed, XML format (equivalent to ABCD) – relatively poor uptake due to problems encountered during both generation and use owing to the complexity of the format. Created seven years ago but there is still little use of this across the community.

·         Simplified XML format containing only the core data (extended form of Dublin Core) – allows for the exchange of the basic attributes associated with records including non-standardised attributes. There has been broad uptake especially for reporting from, and exchange between, standardised collations particularly as it benefits from the vast range of standard tools for dealing with this format.

·         Extendable tab separated text file format – very wide adoption across the data providers, very easy to produce. Use for reporting is limited, standard tools cannot be used directly and data must be loaded into an intermediate format before processing (e.g. Database).

Much of the experience gained over the last seven years is directly applicable to the development of other networks particularly those involving the exchange of observational data. This experience is discussed in the light of parallel developments on an international scale (most notably ABCD and Dublin Core).