There have been worrying attempts in the past year to tamper with the sharing of scientific research results. Each seems indicative of a shift away from the classical principles of science.

There was the review of two H5N1 avian influenza virus studies in ferrets by the US National Science Advisory Board for Biosecurity in December 2011. The board initially recommended that the published studies should “not include the methodological and other details that could enable replication of the experiments by those who would seek to do harm” (see go.nature.com/nywkdy). Fortunately, this attempted ban was later withdrawn.

In two disturbing examples of publication irregularities, a meta-review of biomedical papers in leading journals revealed that key primary data are not always made publicly available (A.A. Alsheik-Ali et al. PLoS ONE 6, e24357; 2011); and Bernardo Huberman alerted the community to the practice of using non-disclosed data from private sources (Nature 482, 308; 2012).

In December 2011 a bill was proposed to the US Congress to reverse the National Institutes of Health policy that all taxpayer-funded research should be freely accessible online (see go.nature.com/uvj68l). The bill's proponents later withdrew their support, but legislative action would have severely limited the diffusion of scientific knowledge.

Science should be available for evaluation by other scientists and for public scrutiny, just as it has been since Galileo's time. It should not be heading for epistemological suicide as a result of vested interests or a creeping loss of awareness of the theory of knowledge.