Sir

I read with interest your leading article1 about the withholding of data from “full and open access”2, which is increasingly at issue across the sciences.

During the past four years or so, my colleagues and I have been developing methods and procedures for the publication of research data in general and ecological data in particular. Our initial motivation came from the work of the Ecological Society of America's FLED (Future-of-Longterm Ecological Data) Committee3 and more recently as part of a National Science Foundation grant supporting the development of a Web-based data management system for ecological analysis and synthesis4. This work continues to involve the society (publisher of Ecology) at the editorial, committee and research collaboration levels because of the unique and influential role professional societies and journals play in the debate about intellectual property rights in data.

There are recurring and fundamental issues limiting ‘full and open access’ to data that are intrinsic to institutionalized scientific research. Examples include the fear of being ‘scooped’ by someone using one's data or inadequate attribution for one's intellectual investment in a research programme resulting in the data, and the relationship this has to academic career advancement. The efforts of Nature and Science to address the issue of restricted access to data is a crucial and significant crack in the cultural mind-set that fosters, permits and even necessitates the withholding of research data.

The next steps along this path require fuller discussion and involvement of the scientific community along with the funding agencies and academic policy-makers. New ideas are needed. As digital library and data repository technology improves and the potential for broader and more rapid dissemination of data increases, consideration needs to be given to new concepts of publication. One approach would be to raise data collections to the status of citeable entities in journals. Academic merit could then be obtained for the development and maintenance of data collections and it would be easier to recognize costs for data maintenance in grant proposals. Without such changes, there will continue to be little incentive for individual scientists to ensure the long-term quality and integrity of their often priceless data.