How is this different from Google searching then copy and pasting from Stackoverflow? Software is hard because of the 20, not the 80.
40% of the code GitHub Copilot users check-in is AI generated and unmodified
61–70 of 129 posts
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#62I'm not surprised. A lot of the code I'll write in a day is trivial, calling helpers, writing test scaffolding, etc. I'm still building cool things the other 60% of time, but this job as we know it comes with a lot of repetition. (esp. if you write unit tests ;-))
The real question is, why we need to write "trivial" code? I think people have lost the plot there. We should be writing better compilers and libraries, instead of using AI that will write more code to deal with. The obsession with code generation instead of creating proper abstractions is a scourge of today's programming. It's a regression in software engineering.
I have made the same point elsewhere to AI enthusiasts friends of mine.
The fact that defining the same task in natural language to an AI is easier than writing it in formal language, means the formal language lacks expressiveness [0], vocabulary, or more likely, both.
But if they are both at same level of expressiveness and vocabulary, formal language wins, because it can spot and prevent ambiguity.
The question is how scalable it is to implement such vocabulary into formal language.
[0] - The line is often blury. How many languages allow to straightforwardly express `clamp m between 1.0 and 2.0`? You can argue this is vocabulary, or this is expressiveness.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#63How is this different from Google searching then copy and pasting from Stackoverflow? Software is hard because of the 20, not the 80.
A. Direct solution vs choosing from 4-5 stack overflow answers. Gives an illusion of doing your work for you.
B. AI can slightly manipulate existing code according to your prose. But I have also seen it fail hilariously to follow basic instructions. Eg: without `All argument constructor` in - but it still generates the all argument constructor. Another day it writes regex but fails to escape some characters properly.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#64There is a lot of annoyance in the boilerplate of a program - "set up a scaffold for this, template that, convert data from this format to that format". The AI tools I have experimented with are good at this and for sure this is useful. But no one can read my mind! If I have to get increasingly specific in what I am telling the AI, and have to formulate my writing style in a way that keeps things very precise, I am e…
I am using Copilot and I sometimes have the feeling it does. About well 20 % of the time. 60 % is quite useful and about 20 % it gets in the way. But these mindreading moments are amazing
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#65Earlier quoted context omitted.
I just started using Copilot and so far, it really does seem to save me time with rote stuff like imports, function definitions, and where I would otherwise be copying / pasting code I just wrote (say, while writing a bunch of tests). Even when it gets those slightly wrong, it still saves me time or at least makes it feel that way. So overall I'm honestly pleasantly surprised with it. It feels similar to when I got a…
To me, depending on the language, it's less useful than coc-nvim (too slow, breaks my tempo, too much added code, sometimes add stuff I did not ask for, or use old, inefficient patterns you only really find on shitty websites), so I still write the 'clever' part with vim, but when I write tests I now use VScode because copilot is really helpful (and it's funnier to use it than to write unit tests that work). For lang…
There is a very good Copilot plugin for nvim written by the legendary tpope himself. It is easy to toggle on and off. I use it alongside coc-nvim every day.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#66I had it on yesterday and had the most miserable afternoon's coding I can remember for a while.
It was a particularly tedious, ugly, and cludgy bit of work that needed doing. Pulling in a data from several very broken sources with horrible naming conventions, patching holes and generally trying to make a sane dataset out of nonsense.
I'm convinced Copilot made the whole process ten times harder than it needed to be. It was constantly pre-empting me with plausible-sounding variable names and merges which led me off down the wrong path a few times.
I flick between having it on and off, and recently I've found I'm much happier with it off. Great for docstrings, boilerplate, and stuff like "recursively traverse this json for [key] and return [value]", but for day to day stuff it's a pain in the arse.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#67So let’s have a test where two or more teams build the same software. One set has AI tools, the other doesn’t. The software definition is provided to both at the same time, and there is a set of tests to see which team produced the software the fastest and produced the highest quality software.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#68I'm not surprised. A lot of the code I'll write in a day is trivial, calling helpers, writing test scaffolding, etc. I'm still building cool things the other 60% of time, but this job as we know it comes with a lot of repetition. (esp. if you write unit tests ;-))
The real question is, why we need to write "trivial" code? I think people have lost the plot there. We should be writing better compilers and libraries, instead of using AI that will write more code to deal with. The obsession with code generation instead of creating proper abstractions is a scourge of today's programming. It's a regression in software engineering.
Sometimes it is just the syntax of your language. We can't all be writing lisp where the syntax is just another abstraction.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#69Aside: am I the only one here that is still writing code by hand and not using AI to autogenerate it?
Because it seems over the past couple years on HN everybody is making great use of Copilot and constantly talking about it, while honestly scaffolding boilerplate is the least of my problems, while thinking, designing and abstracting novel ideas into code is the hard part where AI would fall short and just waste my time.
I still do not get if I'm behind the times or just experienced enough (I've been programming for two decades) that I don't feel the need to use Copilot. It's a bit disconcerting because I feel I need to get out of pure programming work ASAP as everybody else is competing with me with Copilot and GPT and I am feeling an existential risk to my craft and livelihood.
Re: 40% of the code GitHub Copilot users check-in is AI generated and unmodified
#70I'm not surprised. A lot of the code I'll write in a day is trivial, calling helpers, writing test scaffolding, etc. I'm still building cool things the other 60% of time, but this job as we know it comes with a lot of repetition. (esp. if you write unit tests ;-))
The real question is, why we need to write "trivial" code? I think people have lost the plot there. We should be writing better compilers and libraries, instead of using AI that will write more code to deal with. The obsession with code generation instead of creating proper abstractions is a scourge of today's programming. It's a regression in software engineering.
[dead]