Put yourself in the reader’s shoes when creating articles with ChatGPT
Dismissing new tools is a mistake.
You cannot avoid ChatGPT nowadays. It has been claimed that an average chess player paired with an average computer can beat the most sophisticated supercomputer. You aren’t average, and ChatGPT isn’t either, so it’s your chance to dethrone Stephen King.
But ChatGPT keeps giving me cheesy articles whenever I ask it to write one!
Then, the following quote by Nicolas Cole came to mind:
“Remember, the reader is the main character here. And in order to be an effective writer, you must be in service to them” — Nicolas Cole
It made me think: What if I stopped asking ChatGPT to write me an article, and ask it something else instead …
Check the first prompt below. I followed the typical approach asking the Large Language Model to write me a tutorial:
Prompt I: I want to tutorial an article about Python dataclasses, help me by writing it in 100 words or less.
Title: Python Dataclasses Tutorial
Learn how to create clean and concise data-oriented classes in Python using dataclasses.
With the “dataclass” decorator, eliminate boilerplate code and let Python generate essential methods like
__init__,__repr__, and__eq__automatically.Define data fields with type annotations and leverage default values and nested fields.
Discover how dataclasses streamline object initialization, printing, and comparison, making your code more efficient and readable.
Dive into this step-by-step tutorial and harness the power of dataclasses to simplify data management in your Python projects.
I am sorry! Please accept my sincere apologies for the previous cheesiness overload!
Then I tried putting myself in the reader’s shoes and changed the prompt accordingly.
Prompt II: I want to write an program using Python dataclasses, help me by explaining it in 100 words or less.
Here’s an example program using Python…
