What naming practices should you avoid when naming things?

When new concepts emerge as science advances, they are given new names. These new names are given according to various naming conventions, but computer scientist Will Crichton has argued in his blog that 'some naming conventions should be avoided.'

Naming Conventions That Need to Die | Will Crichton
https://willcrichton.net/notes/naming-conventions-that-need-to-die/

Names are important tools for thinking, providing a lightweight means of managing and structuring knowledge, but Crichton argues that traditional naming conventions impede learning and impede progress.

Crichton points out that the first problem is that 'the practice of using inventors' names is most widespread in the fields of science and mathematics.'

For example, there are many mathematical formulas and theorems named after their discoverers, such as the Cauchy-Schwartz inequality , L'Hôpital's rule , and the Riemann hypothesis , but Crichton points out that this is a factor that hinders understanding of people who learn the concepts later. For example, he says that it is preferable to call it 'normal distribution' or 'bell curve' instead of ' Gaussian distribution '.

They also pointed out the problem of numerical naming: for example, Crichton argued that the terms type I error and type II error in statistics should be replaced with the more meaningful terms 'false positive' and 'false negative.'

Crichton also cited naming conventions that use random words, such as those seen in Apache 's Pig , Flink , and Spark . Names that don't provide context can have the effect of excluding people who aren't familiar with the field, especially in technical conversations, he said.

Historical naming can also be a problem. For example, the car and cdr functions in LISP were named after hardware limitations in the 1950s. Crichton argued that these should be replaced with more easily understood terms such as head and tail.

Crichton argues that 'knowledge should be constructed compositionally,' and that it is most efficient to understand concepts by memorizing a small number of core elements and combining them in various ways. He argued that improving naming conventions is not just a technical issue, but an essential challenge for the democratization of knowledge and effective learning.