In a previous post, I drew some parallels between scientific and democratic processes. Another thought that occurred to me is that evidence in science functions somewhat like policy in politics.
On election day, you have a chance to vote for a government with a particular set of policies. Those policies, however, are very liable to change. And people can change their minds about a policy that isn't working, or if a new policy is introduced that they disagree with.
In science, people change their minds as new evidence becomes available.
In neither case is a decision now and for ever, once and for all.
Of course, it's often much harder to show that policy is faulty than it is to show scientific evidence is faulty. To paraphrase Feynman, if your basic understanding of physics of balances and strains and stresses is bad, your building falls down.