Steve Griffith
Posted on July 14, 2023
Learning is repetition.
This is the most important concept to keep in mind while learning how to program.
When you learn any task, from playing guitar, to baking cookies, to building a brick wall, practicing your skill is what you need to do. The same restriction applies to becoming a developer.
When an AI learns how to program, or more accurately, how to generate code that will accomplish the most likely next steps in your program, then it needs to practice too.
The large language models (LLM) that have been built to help tools like Github CodePilot help you write code required huge amounts of code to analyze.
Machine Learning requires millions or billions of repetition to become effective at what they do. If the Machine Learning AI's job is to identify tumors in X-ray images, then it required millions of sample images that have been classified. It required guided practice where it was told whether or not it was classifying images properly.
It required practice.
It's not that one great programmer sat down and wrote an AI program that knows how to write code like they do. Learning requires repetition and practice.
Human Learning
Despite how far our science, technology, engineering, and medicine have come in the last hundred years, our brains are effectively the same as they were 10,000 years ago.
Humans learned how to pass along skills through watching someone skilled, by emulating them, by using their hands to practice those skills. Through repetition and feedback they improved.
This approach to learning still applies today.
Our brains need time to assimilate new information.
There is only so fast that anyone can learn a new skills.
AI tools need repetition too. They actually need more repetition than people. A human can often grasp a concept and be able to infer greater meaning from just one single example. The AI tools just have the advantage of repeating the steps faster.
Guided Learning with AI
There seems to be a sense now with students that either: there is no point in learning how to program because the AI tools will be doing all the programming soon; or that AI is going to make you able to learn how to program in an abruptly quicker time frame.
Neither is really true.
We still use assembly lines to build cars. Those assembly lines still use people. Many of the jobs that are done on the assembly lines are different than they were 100 years ago. There are many robots that do the heavy or monotonous tasks. The humans do the jobs that require nuance. The humans do the tasks that are more supervisory in nature.
The same is true for programming and AI. As developers learn to use the AI tools, it changes their role. The focus of developers can shift from questions like what is the name of the method I need here?
to what is the most intuitive sequence of interface elements to help my users?
.
AI tools can actually increase job satisfaction for developers. Let the AI tools do the heavy lifting, error identification, and the monotonous tasks. Focus on actually building a better application.
AI is not taking development jobs. AI tools are changing the roles of developers.
As for the second point, and the focus of this article, AI cannot force your brain to learn things faster.
What AI tools can do is assimilate the best guidance from a wide range of tutorials and sources. AI tools can help to guide you through a series of steps in your journey to learn how to program.
AI tools can answer questions about concepts when you struggle with them. AI tools can provide alternate explanations when one analogy doesn't work for you. AI tools can explain your own code to you. AI tools can explain generated code to you. AI tools can provide a series of step-by-step instructions to accomplish a task.
AI tools will NOT make your brain learn faster, per se.
AI tools WILL help you to avoid time wasted if you struggle with a topic. They will help you to organize your learning journey.
AI tools will be your personal tutor, always available by your side to answer your questions and ready to provide your next topic to learn.
You will still need to practice.
You will still need to wrap your head around abstract concepts that programming is filled with.
But, just maybe, AI tools will help to keep you on the right path and keep you from getting lost in the woods.
Posted on July 14, 2023
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.