Videotaping experiments?

Today's issue of Nature has an interesting commentary by Timothy D. Clark, advocating videotaping of experiments.  His motivation is to combat scientific fraud.  However, there are broader reasons to consider the idea.  Clark alludes to some, and here is another.  Back when Google Glass came out, I read about a scientist who recorded the execution of her lab protocol simply as a means of documenting what was done.  This can be an extremely valuable supplement to a written protocol, as a video recording is more likely to capture "folk knowledge" within a lab, that nobody thinks to write down.  In other words, such practices could enhance reproducibility.

Of course, videotaping experiments is a long tradition in fluid dynamics, dating back to Henri Benard's films of what we now call von Karman vortex streets, in the first decades of the 20th century.  Today, image analysis methods are often used to extract data from moving images of fluids.

The other point made by Clark's post is that the burden of proof for allegations of misconduct is on the accuser, rather than the accused.  He makes a point about the trust-based nature of the scientific enterprise.  Shifting the burden of proof to the authors not only reduces the likelihood of fraud, but enhances the likelihood of reproducibility.    DTLR endorses Clark's proposal.