[TheOtherHobbes]

>Programming is like that.

No it's not. Making YouTube videos is like that. Knowing how to do social is like that. Being able to look stuff up on Google is like that. Graphic design is like that. (I'm often amazed by the quality of the graphic design and motion graphics on amateur YouTube channels.)

Knowing how to tell when politicians are lying should be like that, but isn't.

Programming is engineering. Not everyone is an engineer. Not everyone can be an engineer, because a lot of people's minds don't work in an engineery kind of way.

I think it's fine to teach kids Scratch and maybe some Python. But I don't think it has anything to do with effective communication - it's just useful practice for basic logical thinking and problem solving.

The kids who are good at it (maybe 10-15%) and enjoy it (maybe 25-50%) can specialise and do the harder stuff.

In any case I don't expect the Web of 2035 to look much like the Web of today, so there should be no expectation that learning any specific language or system is going to be useful later - any more than it made sense to expect everyone would be using BASIC and CP/M in 2000 when we started seeing computers in schools around 1980.

[homarp]

It works the other way around >Graphic design is like that. Graphic design is art. Not everyone is an artist. Not everyone can be an artist, because a lot of people's minds (or "eye) don't work in an artistic kind of way.

But getting the basics of Graphical Design can still help both appreciate a good design and "understand" why "that stuff" feels ugly, and being able to discuss with a graphic designer and talk the "same language"

Same thing with "Programming". No one claims that 100% of the population should understand Rust borrow checker or functional purity

But everyone should understand (and be able to copy-paste some code and "fix" it until it works) the concept of a sequence of instruction, a loop to generate repetitive stuff, the concept of variable (and therefore template and mail merge), a if (and therefore be able to do basic stuff in Excel). So they can talk to engineer and get what they way. And maybe everyone should be able to do FizzBuzz or display the number from 50 to 1 with a loop going from 0 to 49. At least in pseudo-code.

[TheOtherHobbes]

But my point is that a lot of people who aren't trained graphic designers, know nothing about the history or practice of graphic design, and have probably never studied art are able to produce cool designs that are close to the standard of professional work.

And also that being to do this is a huge personal and business benefit.

Being able to code gives you - what? If you're not doing something useful with code - probably professionally - it's not a communicative, practical skill.

Office, much as I hate it, kind of is.

There may be some intangible benefits. But so far as I know there has been no research to suggest that learning to code improves personal, social, academic, or professional outcomes at school.

Meanwhile there's a lot of research to suggest that learning a musical instrument or a second language has obvious measurable benefits.

Obviously I'm not against coding. But I'm definitely against any mythology of coding that suggests it's a key literacy skill - because based on real evidence, there a lot of other skills with a better claim to that.

[homarp]

let me clarify myself. What I see as a "key litteracy skill" is what some calls "Computational Sense" ("A familiarity with the capabilities of computer applications and the ability to easily grasp the difficulty in implementing a computer-based solution. Typically acquired by learning a programming language.") aka https://xkcd.com/1425/

So the "learning to code" is just a mean to an end.

Also I do think that learning to automate the boring stuff (https://automatetheboringstuff.com/ ) improves at least personal (e.g. do your own taxes in Excel, understand AND and OR so you filter email better), social (e.g. ifttt), academic ( e.g. R ), AND professional (e.g. everything) outcomes.

>learning [...] a second language has obvious measurable benefits

and don't you think that the ability to analyze a process and transform it into a sequence of instructions can have measurable benefits too?

You wrote "it's just useful practice for basic logical thinking and problem solving." I agree. It is only that I would remove the "just".

I like what Papert wrote: " debugging is the essence of intellectual activity".

>And also that being to do this is a huge personal and business benefit.

All the "non technical" startup founders looking for/lacking a technical co-founder, you don't think that they would benefit from learning a bit of coding so they can a) estimate what needs to be done (Computational Sense), b) hire the proper people c) specify what is needed ?

Obviously, I am against coding as "learn js in 5 days" but not being able to FazzBazz(2,7). I am for learning what a computer can do.