According to a man named Samuel Arbesman, half of what you and I know is probably wrong.
This does not mean we are just stupid dunderheads, unless our name is Harry Reid, Michael Moore or Will Ferrell. It means that as science progresses, many of the things we learned when we were studying it in school have been found to be incorrect. New evidence trumps old evidence, new data corrects old data, new discoveries put old understandings in a different light, and suddenly dinosaurs are not cold-blooded ancient lizards but warm-blooded ancient birds and Pluto isn't a planet.
This doesn't necessarily apply to all facts. Arbesman's book, The
Half-Life of Facts: Why Everything We Know Has an Expiration
Date, doesn't mean to suggest that the change in designation somehow affected Pluto, for example. It's just that our definition of "planet" changed, and thus the "fact" that there were nine planets in our solar system is now wrong. Since that fact was manufactured, so to speak, as a by-product of a way we defined planets, then a change in the definition created a different fact. The number of planets is a different kind of fact than, say the speed of light in a vacuum. The first depends on definition, but the second depends on observation. The first can change on a vote, but the second only by new experiments that show old experiments were wrong.
Arbesman suggests that we stop trying to memorize these kinds of facts, since we can look up the most recent data if we need to know the information and not be at the mercy of outdated knowledge. We can, he suggests, outsource our memories to "the cloud," the name given to all of the data available online. Although his examination of the half-life of "facts" is interesting, the idea that holding knowledge for yourself should be replaced by relying on the internet is a suggestion that is probably best forgotten. Michael Moore would be more reliable, as one could simply take the position 180 degrees opposite from him and be reassured of being right a good 90 percent of the time.