Live data from Hacker News

Did GitHub Copilot increase my productivity?

trace.yshui.dev

41–50 of 326 posts

Re: Did GitHub Copilot increase my productivity?

#41
I enjoy working with interns, you can see them learn and they are always making new mistakes. I get a return on the effort of training them. They might even convert into full time employees and take some of my workload.

I don't get that feeling from the LLMs. They have about the same skill level of an intern, but they don't _learn_. I can't offload any work to them, and they take the same level of effort to manage.

I'm not in this job to manage interns. I'm in this job to solve problems. Training the intern is a payment by current me for future me.

This isn't like going from books to Internet search or StackOverflow. It doesn't provide an immediate benefit to me, and I don't benefit in the future from my contributions. I'm not seeing the share-alike vibe necessary for scale.

I want tools that make me faster and more efficient. That's how I keep increasing my wage. Maybe if I could _see_ the AI learn from my training? Maybe if I saw benefit from the effort? However, right now, I'm paying for the benefit of training the LLM.

Re: Did GitHub Copilot increase my productivity?

#42
post #28

Earlier quoted context omitted.

Reviewing code is famously harder than writing it. I don’t see how it’s a gain to have to review more code than to write.

It's shorter, you're working directly with the code base you're already thinking about, and it's plugged into it rather than having to start a new mental task when you're looking at a pull request.

When I write code, I usually get a complete understanding of the problem and constraints. I have a picture in my mind of all the edge cases and branching that can happen, which most of the time allows me to write correct code without bugs.

When I review code, I am completely unable to get this mental picture, and I usually miss a lot of issues that I would otherwise have avoided.

Which is why I don't see the point of Copilot.

Re: Did GitHub Copilot increase my productivity?

#43
I feel like this bit is important for this discussion: "...I do not use Copilot for my day job. I use it for my own projects only..."

My guess is the people that don't use Copilot because they don't trust it, or don't like the code it produces, are probably trying to use it in a professional scenario where they have strict coding standards. For hobbyist use or side-projects, Copilot is an incredible time saver. I use it with my Python side-projects, and it's truly amazing how much time it saves me. Some of the most common use-cases where it saves me time:

- Adding docstrings. Just type `"""` after a method name, pause for a second, and Copilot will generate a pretty decent docstring. And usually it will mimic the style you've used to write other docstrings in the file. - Writing tests. This one can be a bit hit-and-miss, but I mostly get good a good starting point, and in the best case it will suggest test scenarios that I might not have thought about. - Creating basic functions/methods. I have terrible memory and usually need to refer to the docs for even fundamental Python code, e.g. reading or writing CSV files, looping over and modifying dictionaries, etc... But with Copilot, I get great results, often just by typing `def open_csv_file` and then pausing for a second to get the method code. If that doesn't work, adding a comment with what you want the function to do before writing the function name will help a lot. - Adding type hints to existing code. Just type ":" and pause for a second, and Copilot will almost always provide the correct type hint. And pressing a space after the closing parentheses in a function, will almost always return the correct return type hint.

Re: Did GitHub Copilot increase my productivity?

#45
I still haven't found any interest in this wave of AIs for coding, mostly because writing code hasn't been a bottleneck for me since I learned vim and started using it as my only tool.

Knowing what to write and how to organize the code is the really difficult part of our job, and I am unable to do this correctly when reviewing stuff written by others (or by an AI).

I have been wondering if people are actually using those to save time writing boilerplate code, in which case it would make sense.

Vim has greatly increased my productivity by allowing me to save time for all repetitive tasks, and it's consistent and reliable (but also super hard to learn).

Maybe it's just a completely different approach in resolving the same problems?

Re: Did GitHub Copilot increase my productivity?

#46
post #29

GitHub touts a 55% improvement in coding speed based on a study conducted in-house that tested the participants’ ability to paste a pre-written prompt and then check the output: https://github.blog/2022-09-07-research-quantifying-github-c... The effectiveness of a Copilot-like tool trialed at FB showed that 8% of code contributed by participants was sourced from suggestions, but the latter study made no promise about…

My personal feeling is that utilising LLM assistance often isn't faster, but it can take less "stamina", tiring me less.

Re: Did GitHub Copilot increase my productivity?

#47
post #45

I still haven't found any interest in this wave of AIs for coding, mostly because writing code hasn't been a bottleneck for me since I learned vim and started using it as my only tool. Knowing what to write and how to organize the code is the really difficult part of our job, and I am unable to do this correctly when reviewing stuff written by others (or by an AI). I have been wondering if people are actually using t…

I don't understand the boilerplate argument because every major framework has had boilerplate generators for over 10+ years.

Re: Did GitHub Copilot increase my productivity?

#48
post #42

Earlier quoted context omitted.

It's shorter, you're working directly with the code base you're already thinking about, and it's plugged into it rather than having to start a new mental task when you're looking at a pull request.

When I write code, I usually get a complete understanding of the problem and constraints. I have a picture in my mind of all the edge cases and branching that can happen, which most of the time allows me to write correct code without bugs. When I review code, I am completely unable to get this mental picture, and I usually miss a lot of issues that I would otherwise have avoided. Which is why I don't see the point of…

When you review pull requests you're unable to get the mental picture. But it's different with copilot because it fits in your mental picture, you can compartmentalize what you're working on, and the copilot output can come in the form of small pieces that are easily understood and slot in the prepared places where you need it.

I agree 100% that reviewing human code is a lot of work but not with copilot IMO because of the immediacy of it all.

Re: Did GitHub Copilot increase my productivity?

#49
post #29

GitHub touts a 55% improvement in coding speed based on a study conducted in-house that tested the participants’ ability to paste a pre-written prompt and then check the output: https://github.blog/2022-09-07-research-quantifying-github-c... The effectiveness of a Copilot-like tool trialed at FB showed that 8% of code contributed by participants was sourced from suggestions, but the latter study made no promise about…

This last sentence is the most important. You can describe your dB schema in words and get laravel migration, models, controllers and if you wanted policies, form requests etc all near perfect. You can get a V1 in 10-20 minutes then go about handling all the actual logic

So rails scaffold from 20 years ago but with a chance of hallucinations?

Re: Did GitHub Copilot increase my productivity?

#50
post #43

I feel like this bit is important for this discussion: "...I do not use Copilot for my day job. I use it for my own projects only..." My guess is the people that don't use Copilot because they don't trust it, or don't like the code it produces, are probably trying to use it in a professional scenario where they have strict coding standards. For hobbyist use or side-projects, Copilot is an incredible time saver. I use…

I only use Copilot for my day job, because they paid for it. When I'm coding my personal projects on nights and weekends I find myself missing Copilot and I'm contemplating paying for it myself. I don't really trust it for everything, but sometimes it surprises me with how much it gets right.

I just can't justify paying $19/month personally for the "Business" plan so they don't use my proprietary code to train their AI.

Post reply on HN