So basically you want to train a generation of copy/paster who don't understand what they are actually doing and who will never actually deploy anything because it is scary.

Hey, at least those of us who can actually hack it will never run out of messes to bill time and a half to clean up!

I strongly disagree, someone who create a web site or learn to program needs to know the basic of the basic of using a computer

it's not about knowing the basics of using a computer, it's about eliminating the possibility of tiny mistakes like 1 typo in a file name causing everything to break, which can put someone off coding forever

> tiny mistakes like 1 typo in a file name causing everything to break, which can put someone off coding forever

One typo in your math or chemistry homework will quite likely give you the wrong final answer, but hopefully that wouldn't "put you off science" forever. Otherwise none of us would be here as I'm sure we have all made hundreds or more.

Are we not expecting students to be resilient and solve problems anymore? One of the Common Core standards that used to be on the classroom wall was "persevere in problem solving." I say it's great for students to make a few easy mistakes like forgetting a semicolon to get in the habit of reading error messages and troubleshooting while they are still straightforward to fix.

The important context here is that this book is not "HTML for programmers," it's "HTML for people who are probably pretty damn scared about the word HTML right now"

In this case, the single most important thing is *early successes*. Kids spend years learning about the number line and what is the difference between + and - before they ever do 2+2=4. Or if they learn 2+2=4 first, it's just some abstract syllables they were taught to parrot, and they probably don't understand.

For a new programmer, who is ALREADY SCARED OF PROGRAMMING, the single most important thing is early successes. If they can make something work on their first attempt, without realtime help from a friend fixing their mistakes, they are SO much more likely to have the needed self-confidence to keep learning.

For a concrete example, every time I teach regular expressions to people I say, `cat` is a regular expression! Let's search for `cat` in `catastrophe` and turn "regex" mode on! Congrats, you have now written and used your first ever regex!! And this goes over SO well. It's SO much better than trying to start out with a symbol, because I give them an initial win that they achieved and that they were able to do. And if they get stuck later, they can always go back to knowing that `cat` is a regular expression and search for `cat` in `catastrophe`. And if it doesn't work, there's a different problem.

In other words, not only is giving people an early success like this good motivation, it's also teaching them the negative control that they'll use for the rest of their programming careers, even if they don't know it.

"Make a file on your computer" is not useful by itself. It's not a negative control. It's not an early success. It can be learned later, once you have the other things.

If making one mistake is enough to put one off forever then they’re really not going to like programming which is just a sequence of writing bugs and fixing bugs. :)

I'm sorry, but I learned on a garbage Windows 98 machine when I was like... seven, and I actually got quite far without someone padding the entire computer and handwringing about f*cking _typos_. My god, what happened to this industry? Yes, learning is hard! Sometimes you make mistakes! That is how you learn! This is why my coworkers submit garbage pull requests full of obvious errors, because they've been coddled to death and never allowed to accidentally typo a directory name. Christ...

Should we get rid of the keyboard, too? After all, it's so complicated.