Costs and benefits of Web access to museum data

Abstract

Brooke’s1xWhy museums matter. Brooke, M.de L. Trends Ecol. Evol. 2000; 15: 136–137Abstract | Full Text | Full Text PDF | Scopus (16)See all References News & Comment alluded to the concerns museum curators have about unrestricted Web access to biological specimen data – the objective of several competing international bioinformatic networks. All parties agree that museum specimen data have many scientific and commercial applications, but the costs and benefits of open Web access remain to be reconciled.First, relatively few specimens have been computerized (e.g. only 13% of 2 350 000 skins in the three largest ornithological museums in Tring, New York, USA and Washington, USA). At prevailing costs (US$1.50–US$2.00/ specimen in Washington), computerizing the remaining 87% of the specimens would cost between US$3 200 000 and US$4 200 000; however, accurate georeferencing might double or triple the cost. External pressure for computerization is increasing, but who will pay? Museum funding continues to decline and many institutions survive on starvation budgets; thus computerization is a lower priority than other urgent infrastructure needs. And, if such funds were available, wouldn’t they be better spent on new biodiversity surveys?Brooke mused that museums ‘should accept that, by making their data available, they are providing an immensely valuable service to the world community and use that to leverage more core funding from national governments’. If past experience is a guide, the ardent courtship of museums and promises of financial support by bioinformatics advocates would probably vanish as soon as the specimen data have been extracted.The fundamental problem is not free transfer of museum data, but to whom data are transferred and for what purpose. Free access to museum specimens and data has long been provided to bona fide researchers. However, before major museums will permit sensitive data fields (e.g. collector, date, specific locality and reproductive data) for entire collections to be posted on the Web, bioinformatics networks must provide legally binding safeguards and provisions. These safeguards include: (1) intellectual property rights; (2) commercial use and licensing; (3) appropriate acknowledgement in the publication; (4) endangered species data; and (5) privacy rights of collectors and researchers. At present, the major networks rely on unsecured disclaimers to prevent data miners from systematically downloading, repackaging and selling museum data, from transferring data to third parties, and from using compilations of raw museum data to leverage grants and contracts in direct competition with museums. Reliance on the honor system to protect the intellectual property rights of museums and researchers is naive. Moreover, it is virtually impossible to compel Web database users to acknowledge data sources appropriately (e.g. Ref. 2xConservation of ecological niches in evolutionary time. Peterson, A.T. et al. Science. 1999; 285: 1265–1267Crossref | PubMed | Scopus (802)See all ReferencesRef. 2). Unfiltered data might also be used to locate and exploit commercially valuable or endangered species, and might provide the means for anticollecting zealots to interfere with ongoing research programs. Once released, the digital genie can never be put back in the bottle.Until flaws in network safeguards are fixed, museum curators can better facilitate free access to museum resources, and protect the interests of museums and researchers, by handling data requests on a case-by-case basis – the same as requests for specimen use.