The analogy is quite strong, here - perhaps more than the author realizes. I get the basic point about not expecting just pick up a book or two and crash the gates, overnight. But there's something fundamentally disheartening about a statement whose core message boils down to: "Don't try".

It's so important that even learning to pseudo code is going to take you a very very long way.

Code is IMO language. It's the language if you'd like of the future, how cliché that might sound.

Understanding even the basics of code (as of music) is almost of value way beyond becoming an expert in the field.

Music is actually a good example to discuss because, while you're playing piano, you could've been playing tennis. While you're playing tennis, you could've been in art class. While you were in art class, you could've been hiking with friends. While you were hiking with friends, you could've been doing more math. While you were doing math, you could've been reading literature. These things are not for everyone, and that an activity has value is absolutely inadequate in proposing a global prescription for children.

I can see a place for civics, economics, exercise, biology, history, math, science, and English as global prescriptions for young people, but programming is far less obvious.

You literally can't find a more versatile tool for that. So the whole idea that people shouldn't learn to code is rooted in either some deeply misguided understanding of how humans grow intellectually or som deeply juvenile idea that if you aren't ready to put in the work to be a real coder you shouldn't get started to begin with.

It doesn't matter if it's not for everyone, people will find that out when they try it out. Some of them will become professional developers many wont, but most will develop an understanding and insight into the very thing that is disrupting their lives in an increasing maner and perhaps use it to further develop other fields.

So the question is not "should we have dance" in schools, but, "What do we cut? What do we downsize? How about exercise or history?". Alternative, "instead of programming, how about statistics and personal finance?" These are far more interesting questions, because these questions acknowledge constraints and weighing. If we didn't have to exercise constrained thinking, we mind as well say, "let's have it all".

Saying that programming is at the intersection of so many subjects is unrealistic, and the same could be said about math, physics, or statistics, and I'd say that statistics and personal finance are far better candidates if you had to somehow squeeze one more thing into school at the cost of something else.

The reason why software exists at some intersection with another discipline is almost always statistics. Statistics, then, is the most flexible gateway to numerous disciplines, and software is the tangential detail to statistics. The scientist exercises experimental modelling, collects the data, and inputs it into an excel sheet, and imports it into the university software suite. And this is the right level of involvement between a scientist and software. The scientist is rightly not interested in the details of software. Nor the physicist, or the student of history, literature, or the exercise enthusiast or the musician. These people are all very justified in throwing money at the problem for a prepared solution, because software is the detail.

You are trying to build up a false contradiction. Different choices brings people forward in different ways but it can never be a bad thing to learn something even if you don't end up becoming a professional. The cost you are talking about is simply not there with programming. Your ability to access almost any field greatly increases with the ability to program, especially in a world where the collective information accumulated to almost double every second year.

You can use programming to explore all those other fields. Whether music, personal finance and so on.

Saying that it's unrealistic that programming is at the interection of many subjects makes absolutely no sense.

Programming embeds statistics not the other way around.

It's the one thing that can be used to simulate almost all other phenomena expect the experience itself. And so the cost would be NOT to understand at least the fundamentals of programming.

If you disagree with this then that's your right, but to claim that someone shouldn't be learning programming is just as absurd as claiming no one should learn music or football just because there is a cost.

That "cost" is perspective and perspective is always valuable.