Ask HN: Anyone else struggle with how to learn coding in the AI era?

I truly believe, that even before Vibe Coding the amount of abstractions one is developing against has been in the way of learning programming and feeling good about it.

You do React + Redux or any other framework and feel like a lot of decision have been made for you without grasping the actual reasoning for these decisions.

The best learners I have encountered and for a year, I am trying to implement:

Learn the platform you develop for on a side project. You develop for the web and more on the programming side: Learn JS for Web and HTML. You will encounter state management, animations, events, DOM manipulation etc.. Solve them without libraries, first.

Imho (in my harsh opinion, in this case) unless you're a prodigy you're probably not a very good programmer at this point. Now granted people learn more quickly outside of university I think because it tends to be more focused, but I don't think I know anyone that I would have called a good programmer a year in.

But feel free to call yourself a programmer, I'm not going to gatekeep it :)

2 things have not changed since the advent of AI, and never will change. The first is the calculus of computation. The second is the engineering of computation. All coding is just some interesting combination of these two forms of logic.

You can learn both of these quickly to a deep level with only 2 books. For the calculus of computation, read "Structure and Interpretation of Computer Programs" (SICP) by Abelman and Sussman. It is available for free.[0] If you understand all of this book, you understand all of the fundamentals of computer science. Every program you write from now on will be understandable to you, with enough persistence. But most importantly you will be able to think in computer programs by second nature, and communicate in this language. And when you talk to AIs in this language, they become exceedingly precise and powerful, because you lost the ambiguity in how you are conceptualizing the program.

For the engineering of computation, read "The Elements of Computing Systems" by Nisan and Schocken.[1] An abridged version of this book is available for free in the form of the Nand2Tetris course. In this course you will start with an imminently simple digital construct, and use it to build step by step a full working computer that can run Tetris. You could even write a Lisp from SICP on this computer, and pretty easily too as you'll see in SICP itself! Once you have completed both books you'll have met in the middle between abstract computer science and concrete computer science: coding.

Just like in your math class, you can see that one side of a right triangle is always longer than the others, but you cannot understand how or why or explain it or work with it until you can comprehend some simple theorems and functions, you cannot truly compose computer programs until you can speak the language of computer science. It used to be that you could make a career by copying code you saw online, patching bits and piece together to create basically working code. But that era is over. AI reads and writes and searches millions of times faster than you. But still only humans are capable of new compositions. But in order to create these new compositions you have to be able to speak a mutual language that you and the AI can understand. That language is computer science, and it hasn't changed since time began and it won't change in 10, 100 or 1,000 years from now when AI is capable of doing anything and everything better than we can. So if you want to stop struggling and start creating new and exciting things with computers, read these 2 books!

[0] https://mitp-content-server.mit.edu/books/content/sectbyfn/b...

[1] https://www.nand2tetris.org

> AI has changed nothing in terms of learning to program

In terms of what you should be doing when you learn to program, I fully agree.

In terms of the effects AI has on the activity of learning to program, I think it has: it has made it very tempting (and affordable - so far) to just make the AI build and even adapt the simple stuff for you that you'd otherwise be building and adapting by yourself. I suppose it can even give you the false feeling you understand the stuff it has built for you by reading the generated code. But this makes you never go through the critical learning steps (trial and error, hard thinking about the thing, notice what you are missing, etc).

We already had the possibility to run web searches and copy paste publicly available stuff, but I think that this came with more friction, and the automated adaptation aspect was not there, you've had to do it by yourself. I think Gen AI has made it way easier to be lazy in the learning and it's a trap.

But from the rest of your comment it seems we mostly agree.

I'd like to draw a parallel to carpentry:

A carpenter uses tools to shape wood into furniture. Each tool in the toolbox has different uses, but some are more efficient than others. For example, a table saw lets the carpenter cut more quickly and accurately than a hand saw. Nobody would say "that's not a real carpenter, he cheats by using a table saw".

A carpenter can also have an assistant (and I'm specifically not talking about an apprentice) who can help with certain tasks. The assistant might be trained by someone else and know how to perform complex tasks. When the carpenter builds something with the assistants help, is that considered a team effort? Does the carpenter need to take responsibility for the assistants mistakes, or the trainer? Who gets credit for the work?

I don't have answers for these questions, but I think the parallel to software is straightforward: we have a new tool (assistant) that's available, and we're trying to use it effectively. Perhaps it's going to replace some of our older tools, and that's a good thing! Some of us will be lazy and offload everything to it, and that's bad.

I do think that learning the fundamentals is as necessary as ever, and AI is a great tool for that as well.

(Disclaimer: I've been programming for about 15 years, and haven't integrated AI into my workflow yet.)