You recently conducted an online popularity poll of three proposals competing for selection as the next NASA Discovery Program mission (Nature http://doi.org/h79; 2012). In my view, the concept and execution of this poll demeans Nature and belittles what is at stake.

Worse, there are indications that the poll could have been manipulated. Voting for one particular mission occurred in a large burst on a single day. It is immaterial whether this was caused by the mission teams enlisting many supporters to vote quickly, or by people who worked out an easy way to vote multiple times. The point is that the results are not meaningful.

Popularity contests are not the way to choose among scientific alternatives. Although public interest needs to be taken into account when spending taxpayers' money, selecting a mission should ultimately depend on its scientific merit and technical feasibility.

NASA's missions have a track record of exciting the public anyway, with web hits for different missions leading to server saturation during key events. The likely effectiveness of each mission's outreach programme needs to be evaluated by looking carefully at the large, detailed proposals submitted by each mission team.