You can learn quite a lot from finding out what a measurement doesn’t tell you. A study of home energy improvements can count new installations without finding a change in energy use. A coin can have the same width all the way round without being round. I doubt the coin has anything to teach a council, but it’s worth a look in its own right.
What happens when local people get local figures?
During the 2022 energy crisis, researchers randomly selected local authorities in England and Wales for an experiment. They sent councillors, officers, and local media estimates of what rising energy prices meant for households in their area, and what different improvements might save. This was information about their own patch, not another national figure to divide by the population.
In a working paper published on 13 November, Menna Bishop and colleagues report a 7% increase in retrofit installations, mostly solar panels, without a detectable change in energy consumption. Installations were concentrated near councillors’ homes. That fits the idea that information travelled through local relationships, though it doesn’t establish who passed it on or how anyone made a decision. The full paper is available from Warwick.
I’d like to know who heard enough to act and who didn’t. An installation is easy to count, which is useful. It still leaves the questions of who could afford one and what difference it made.
What gets passed on when people learn a language?
In a 2004 study of Nicaraguan Sign Language, Ann Senghas, Sotaro Kita, and Aslı Özyürek found that younger cohorts were more likely to express the way something moved separately from the path it took. Think of rolling down a slope: ‘rolling’ and ‘down’ can be expressed separately and recombined with other movements and paths. The children were doing more than copying the signing they encountered.
A separate lab experiment by Simon Kirby, Hannah Cornish, and Kenny Smith passed invented labels along chains of learners. Making the labels easier to pass on sometimes meant using one label for several different pictures. When the researchers stopped those ambiguous pairings passing to the next learner, labels developed reusable parts for things such as colour, shape, and movement. It’s a different kind of evidence from the language in Nicaragua. The lab study lets you see what changes when a distinction has to survive being handed on.
Why a 50p piece isn’t round
The Royal Mint’s account of the 50p makes more sense if you picture two parallel rulers touching opposite sides of the coin. Turn the coin and the gap needn’t change. That’s what ‘constant width’ means. The seven curved sides let the coin feel different from a circular one without changing the measurement a machine can use.
There’s a further surprise. As Barbier showed in 1860, two shapes with the same constant width have the same perimeter. A circle and a rounded Reuleaux triangle can therefore take exactly the same length of string around the edge while enclosing different areas. If you’ve got a 50p to hand, it’s a pleasing thing to look at again.