SwiftUI Charts Mastery book

  • Jun 1, 2025

SwiftUI Charts Mastery - Book Update 5

I wanted to provide everyone with an update on the book's progress.

Book Chapter Writing Progress

✅ Concepts

✅ Your First Charts

✅ Modeling Your Data

✅ Chart Marks

✅ Charts

✅ Common Modifiers

✅ Chart Customizations

✅ Chart Axes

✅ Axis Labels

✅ Annotations

✅ Chart Legends

✅ Scrolling Charts

✅ Animating Charts

✅ Chart Interactions

A little over 250 pages written so far.

Book Feedback

I have now reviewed and applied all the feedback I received from my beta readers.

Thanks to everyone who provided feedback!

Incorporating AI

AI is now becoming an important part of software development, including SwiftUI development.

So I am adding sample prompts at the end of all chapters to help you become more familiar with how to prompt and have AI get some of your work done for you.

AI Prompts

This is almost complete, and I should be releasing the book soon after.

Is AI perfect?

Heck no.

It will get things wrong.

It will use deprecated or outdated code.

It will make up functions and modifiers that do not exist in SwiftUI (these are called "hallucinations").

That is why I also provide some tips and warnings with prompts to help guide you in using AI to improve results.

So why use AI if it's not perfect?

AI will only get better.

AI is not going away.

Right now, it's more correct than it is incorrect.

Reading the AI prompts that I include is meant to guide you and teach you to start thinking in terms of prompts to produce what you need.

If I can use AI, why do I need your book?

Ha ha, great question!

  1. You can't prompt AI to do something you don't know exists.

  2. If you want an advantage over your competitors who only use AI, the book can show you what else is possible outside of common AI responses.

  3. When AI gets something wrong, the book can help you fix it, as all examples actually work.

  4. This visual book will give you the "Picture Superiority Effect" and allow you to recall these images (screenshots) 60,000 times faster than code. This is a known phenomenon in which people can recall images faster than written or verbal information.

  5. AI won't provide you with tips, tricks, and options when you prompt it to do one thing. It's more literal in fulfilling your prompt request.

  6. A common problem with AI is that it is trained on all versions of Swift and SwiftUI. This includes outdated code.
    AI has trouble separating outdated (deprecated) code from current code.
    And depending on the AI model and how recently Apple has updated its documentation (and how complete that documentation is), it's common for AI to not yet have been trained on the new documentation or have been trained on Apple's incomplete documentation.

  7. Apple's own documentation and example code can be wrong.
    I have opened up many Feedback Requests for their incorrect documentation and non-working code examples.
    When AI is trained on incorrect documentation, it gives incorrect responses and perpetuates the problem.
    AI is only as correct as the data it is trained on (garbage in, garbage out).

  8. Spiral of death. AI can generate breaking code. Then, when you try to use AI to debug it, it just gets worse, and worse, and worse, until you get to the point where you end up with a pile of messy code and end up starting all over.
    In this case, it's crucial to have a good working foundation and then use AI to make small updates where you know how it should work and you're just saving time with the code generation.

  9. No training. AI can't give you answers for things it hasn't been trained on. I've already run into this. I think because not only is Apple's documentation lacking, but there just aren't a lot of examples for some of these things on the web, especially for the interaction and animation chapters. Some of those topics I had to figure out on my own to make them work.

Summary

AI is great at generating code you already know how to do.

Competent developers can quickly look at AI-generated code and tell if:

  1. This is exactly what I wanted.

  2. This is not right.

  3. This is using deprecated code.

  4. This will work.

  5. This will not work.

  6. AI is hallucinating.

  7. This is a good practice.

  8. This is a bad practice.

A non-competent developer lacks judgment and simply accepts the response provided by AI, hoping for the best.

So any current books, blogs, videos, or classes are still reliable ways to establish your base of knowledge and will help you build the judgment needed to be able to tell the good responses from the hallucinations. 😃

What's Next?

After the book is complete, I will work on the sales page.

Once that is complete, I'll send out a notification that the sales page is open to start accepting orders!