Software engineer (now upper mgmt) with 25+ years experience, and I'm going to disagree on a couple points.
First, people often assume programming is more closely related to math, when it's really about structured thinking and the ability to break down problems into very, very small components. Just because you are good at higher math does not mean you're going to be good at software, especially since geometry and logic are more closely related to programming than calculus is. It is perfectly possible to teach someone to program well without them knowing any more math than basic arithmetic -- my son learned the basics well as a youngster long before he encountered algebra.
Second, the important correlation between math scores and programming competitiveness isn't the mean test score, but rather the number of people who fall into the highest tier. If you improve your math education to bring up a lot of people from dreadful to adequate, your mean test score will go up... but someone who requires education to get to adequate is lacking the right wiring to make a career out of programming. They're not a naturally logical thinker, or problem solving doesn't scratch the right itch for them. If you want to increase the number of programmers, you want to identify the people who are already on the right side of the bell curve, push them farther to the right, and give them early exposure specifically to programming and software engineering concepts.
There's a lot of overlap in people who gravitate to both programming and math, but there's also very strong overlap with musical ability and ability to speak multiple languages. They lean toward programming because that's how their brains work, not because the math teaches them programming. We wouldn't try to breed more programmers by improving our music education curriculums, and we acknowledge that generic music education doesn't automatically increase the number of highly talented singers. Software is more similar to that IMO.
Interesting—these are also the same kinds of skills needed for cryptanalysis—musicians tend to be solid codebreakers and creators. Maybe we should start looking at what high school athletic programs we can reduce and how we can expand music and early coding programs.
So interesting. My son, the one with autism, is a gifted programmer. He's good at music and languages, too. Cybersecurity will probably be his niche because his IQ is in the top percentile for pattern recognition.
But back to my newsletter... If what you say is true -- improving math education across the country will not yield more computer programmers -- you do have a solution: 1. provide better training in programming and software engineering to students at an earlier age and 2. provide more support for higher achievers.
That's an excellent plan. That plan, too, would require a centralized direction and additional funding, both of which would have to come from the federal government.
I enjoyed your article. Perhaps it is due partially to poorly designed spiral math curricula.
Software engineer (now upper mgmt) with 25+ years experience, and I'm going to disagree on a couple points.
First, people often assume programming is more closely related to math, when it's really about structured thinking and the ability to break down problems into very, very small components. Just because you are good at higher math does not mean you're going to be good at software, especially since geometry and logic are more closely related to programming than calculus is. It is perfectly possible to teach someone to program well without them knowing any more math than basic arithmetic -- my son learned the basics well as a youngster long before he encountered algebra.
Second, the important correlation between math scores and programming competitiveness isn't the mean test score, but rather the number of people who fall into the highest tier. If you improve your math education to bring up a lot of people from dreadful to adequate, your mean test score will go up... but someone who requires education to get to adequate is lacking the right wiring to make a career out of programming. They're not a naturally logical thinker, or problem solving doesn't scratch the right itch for them. If you want to increase the number of programmers, you want to identify the people who are already on the right side of the bell curve, push them farther to the right, and give them early exposure specifically to programming and software engineering concepts.
There's a lot of overlap in people who gravitate to both programming and math, but there's also very strong overlap with musical ability and ability to speak multiple languages. They lean toward programming because that's how their brains work, not because the math teaches them programming. We wouldn't try to breed more programmers by improving our music education curriculums, and we acknowledge that generic music education doesn't automatically increase the number of highly talented singers. Software is more similar to that IMO.
Interesting—these are also the same kinds of skills needed for cryptanalysis—musicians tend to be solid codebreakers and creators. Maybe we should start looking at what high school athletic programs we can reduce and how we can expand music and early coding programs.
You know that I love that idea.
So interesting. My son, the one with autism, is a gifted programmer. He's good at music and languages, too. Cybersecurity will probably be his niche because his IQ is in the top percentile for pattern recognition.
But back to my newsletter... If what you say is true -- improving math education across the country will not yield more computer programmers -- you do have a solution: 1. provide better training in programming and software engineering to students at an earlier age and 2. provide more support for higher achievers.
That's an excellent plan. That plan, too, would require a centralized direction and additional funding, both of which would have to come from the federal government.