PRESS RELEASE – International Council for Science endorses open access to scientific record; cautions against misuse of metrics — ICSU

"In a strong show of support for open access to the scientific record, the Assembly, which unites representatives of 120 national scientific academies and 31 international scientific unions, today voted for the statement which stakes out 5 key goals for open access, and offers 12 recommendations that pave the road for attaining them. 'Open Access is a key mechanism to support the development of science and of vital importance to all scientists both young and old,' said Prof. John Ball, who led the ICSU working group that developed the statement. 'It is a powerful tool for creating and validating knowledge, and for supporting science as a public good, and not as something carried out behind closed doors,' he added. The five goals in the statement assert that access to the scientific record should be free of financial barriers for any researcher to contribute to; free of financial barriers for any user to access immediately on publication; made available without restriction on reuse for any purpose, subject to proper attribution; quality-assured and published in a timely manner; and archived and made available in perpetuity. The statement also makes twelve recommendations for achieving these goals, including recommendations on metrics, stating that these, when used as an aid to the evaluation of research and researchers, should help promote open access and open science. It also cautions that metrics should be regarded as an aid, and not a substitute, for good decision-making. They should not normally be used in isolation to assess the performance of researchers, to determine appointments, or to distribute funds to individuals or research groups, for which it says expert review is indispensable. The Council's position takes account of the specific situation related to research data, asserting that publishers should require authors to provide explicit references to the datasets underlying published research. They also should require clear assurances that these datasets are deposited and available in trusted and sustainable digital repositories. Citing datasets in reference lists using an accepted standard format should be considered the norm. The statement also suggests that terms of contracts governing the purchase of scientific periodicals and databases by libraries serving universities and research establishments should be publicly accessible."