To know that we know what we know, and to know that we do not know what we do not know, that is true knowledge.

NICOLAUS COPERNICUS (1473-1543)

The Half-Life Of Facts

What we think we know changes over time. Things once accepted as true are shown to be plain wrong. As most scientific theories of the past have since been disproven, it is arguable that much of today’s orthodoxy will also turn out, in due course, to be flawed. This is called ‘pessimistic induction’.
 
But what's really interesting is that studies of the frequency of citations of scientific papers show they become obsolete at a predictable rate. Harvard mathematician Samuel Arbesman calls this 'the half-life of facts'. Just as with radioactive decay, you can’t tell when any one 'fact' will reach its expiry date, but you can predict how long it will take for half the facts in any discipline to do so.
 
In medicine, for example, 'truth' seems to have a 45-year half-life. Some medical schools teach students that, within a few years, half of what they’ve been taught will be wrong – they just don’t know which half. In mathematics, the rate of decay is much slower: very few accepted mathematical proofs get disproved.
 
Let’s assume, for argument’s sake, that the half-life of a QI fact is 10 years, which would put us on a par with physics. That means about 7% of what we say in each episode will turn out to be wrong by the time the show is given its first repeat in a year’s time. This is one of the reasons that we run the QI Quibble blog, on which we post clarifications and corrections for mistakes on the show, but also updates about information which has changed since each episode was recorded.
 
 
Here are a few examples of facts from QI's history 'decaying':
 
In Series I we said that nobody knows how to tell the age of a lobster. Since then, Canadian scientists have worked out how: it involves dissecting their eye-stalks and counting the rings.
 
In Series G we said that giraffes' necks may have evolved for fighting each other, because a long neck gives them more momentum to swing their skulls at each other. A recent study has found very little evidence for this hypothesis.
 
In Series A we said the best-endowed millipedes had 710 legs. Soon afterwards, a millipede with 750 legs turned up. (They still haven’t found one with 1000, though).

Perhaps the most notorious instance is when, Series A, we said the earth had two moons. In Series B, we said there were either five or one, but definitely not two. These ideas were novel at the time (and quite controversial) but, as of Series K, scientists now think the Earth has thousands of moons. At least one is the size of a washing machine, with a thousand more as big as a basketball. NASA calls them ‘mini-moons’ or Temporarily Captured Objects (TCOs). Most stay in orbit for about a year before spinning back into space or burning up in Earth's atmosphere. Computer simulations of the paths of the 10 million asteroids in the Solar System suggest 18,000 of them are going round the earth at any one time. Only one has so far been confirmed by direct observation. RH120 is about the size of a car and made four orbits of the Earth in 2006-7.