> I'm quite keen to use it more in my side projects, I just don't really see the time pay off to invest learning it…
I think using it for side projects is where it shines, especially if you’re talking about coding side projects.
In my N=1 experience with coding, it can save me a ton of time by iteratively providing code and saving me time looking up documentation to find correct functions and syntax. ChatGPT frequently shows me functions that I wasn’t aware existed. It also can give me cleaner and/or more idiomatic code.
So a function or class that may have previously taken me 45 - 60 minutes to build, now takes 5-10 minutes. This is a huge productivity increase. It’s not some magic pill that builds whole applications, but it’s like having an experienced (but not expert) programmer by your side with unlimited patience for your questions. (This also applies to non-programming tasks as well.)
And it hasn’t taken me any time investment in learning it. I just have an iterative conversation with it, including asking it clarifying questions and requesting it to modify the code it provides.
Are you experiencing something different than what I’ve described?
My experience is that chatgpt consistently showed me API functions that don’t exist. But I tend to do very obscure things in embedded, that googling for can be tricky or impossible. At first I had high hopes, because the shape of the code is good, and the logic looks good, but when I go to look up details of the parts I actually need help with, they’re just made up functions, like placeholders I might write until I can go back and implement them.
Where it’s shined for me was writing code where I couldn’t be bothered to put together ideas from several distinct tutorials. It’s fantastic for writing code that you can basically google for, even if you have to read several disperate pages to get a clear picture of what to do.
> was writing code where I couldn’t be bothered to put together ideas from several distinct tutorials. It’s fantastic for writing code that you can basically google for, even if you have to read several disperate pages to get a clear picture of what to do.
I think this description of yours nails it.
Soon, though, you’ll be able to feed it a corpus of documentation and/or library code and question it based on those. That might help your use case.
I've been writing an open gl based game engine for scratch, but I'm using c#, which means I'm using a lot of bindings for libs that only really exist in c++. Because of this chat gpt seems to get stuck on functions that don't exist in my context
Side projects benefit greatly from ChatGPT because usually:
- the projects are small in scale, meaning ChatGPT needs less context, and stereotypical solutions probably work quite well
- you're not a domain expert (as opposed to your paying job, which hopefully you are an expert in the niche there), this means relatively speaking ChatGPT can help you more, whereas if you're a domain expert, you probably outperform ChatGPT anyway.
That said, ChatGPT often replaced Googling and stackoverflow when I need to look up how some things are done. The benefit is that I get to ask very specific questions and get to ask follow up questions.
I think using it for side projects is where it shines, especially if you’re talking about coding side projects.
In my N=1 experience with coding, it can save me a ton of time by iteratively providing code and saving me time looking up documentation to find correct functions and syntax. ChatGPT frequently shows me functions that I wasn’t aware existed. It also can give me cleaner and/or more idiomatic code.
So a function or class that may have previously taken me 45 - 60 minutes to build, now takes 5-10 minutes. This is a huge productivity increase. It’s not some magic pill that builds whole applications, but it’s like having an experienced (but not expert) programmer by your side with unlimited patience for your questions. (This also applies to non-programming tasks as well.)
And it hasn’t taken me any time investment in learning it. I just have an iterative conversation with it, including asking it clarifying questions and requesting it to modify the code it provides.
Are you experiencing something different than what I’ve described?