Most people think that learning computer code requires solid math skills. In reality, coding makes much more use of linguistic qualities, which favors people who speak foreign languages. But other experts believe that doing without math leads to poor quality code.
Do the terms of HTML , CSS , Python, PHP or JavaScript seem totally nebulous to you? You are probably part of the 20% of French people who admit that they do not know exactly what the computer code covers. However, three out of four French people also believe that it is essential to know computer coding to train for the jobs of tomorrow, according to a Kantar study for Orange. Despite the myriad of online courses and tutorials available on the Internet , training provided by public or private organizations, many of us still believe that coding is reserved for maths.
Gifted in languages, gifted in computer code
“ Anyone can learn to code. And everyone should try, “said Bill Gates in 2016. In 2017, Apple CEO Tim Cook also said that learning to code should be more important than learning English for schoolchildren. Statements confirmed by a new study from the University of Washington, which show that a natural ability to learn languages is a better predictor of learning programming than basic knowledge of math or calculus. The researchers passed a battery of tests to around thirty adults who had never learned to code in order to assess their cognitive skills in different areas (mathematical, linguistic,problems, memory.
They then took a series of lessons and online quizzes in Python, one of the most common computer languages . Those who learned Python more quickly and with greater accuracy are those who have a mixture of strong problem-solving and language skills, the study found, published in the journal Scientific Reports . In other words, people gifted with languages would be the most apt to learn the code.
The brain activity of each individual determining for learning the code
The researchers also measured the brain activity of participants at rest. Chantel Prat, professor of psychology at the University of Washington and main author of the article, had shown in previous experiments that neural activity at rest alone explains 60% of the variability in the speed of learning. a foreign language (in this case French). The same observation is valid for the code. ” Language ability, fluent reasoning and working memory , as well as resting brain activity are all more important predictors of learning in Python than arithmetic, which accounts for just 2% of the differences between people “,
Learning code without math, a very bad approach?
An opinion that is not shared by everyone. For Leslie Lamport, winner of the 2013 Turing Prize , ” Mathematical thinking is a necessary step to specify the algorithm of computer programs, because it helps programmers to clarify their thinking and make programs more efficient, ” explained the computer researcher. at a conference organized by the Singapore Management University (SMU) in January 2020. ” Writing code without knowing what an algorithm is, is like building a building without a plan “.
The American researcher developed a computer language called TLA + and based on mathematics, which notably made it possible to divide by ten the size of a system ofused by the Rosetta space probe . ” In addition to being more efficient, the mathematical approach makes it easier to debug ,” he says. Many people are put off by mathematics, but it’s not that complicated as it sounds. Why is “Two plus two equals four” considered simple, but a logical operation as “an element of” is difficult for most people to understand? “