Re:40% of the code GitHub Copilot users check-in is AI gener...
by nathansherburn on Wednesday March 15, 2023 @03:00AM
(#35164834)
I found during my preview months ago that it would often write bugs that took me longer to correct than if I had written it myself. Has it gotten better?
Also in the situation where it suggests x[i+1] and it's wrong, doesn't that annoy you?
In both cases I'm sure this is just temporary while LLMs get better. But my experience was they were _just_ a little too annoying last time I used them.
Re:40% of the code GitHub Copilot users check-in is AI gener...
by bananapub on Wednesday March 15, 2023 @04:34AM
(#35165435)
> Has it gotten better?
yes
Re:40% of the code GitHub Copilot users check-in is AI gener...
by markasoftware on Wednesday March 15, 2023 @03:26PM
(#35173466)
Indeed, it introduces subtle bugs related to indexing pretty often. At one point I had it autocomplete maybe 15 lines of code, and I trusted it because the code was all pretty easy and I had written a descriptive comment, which tends to help out Copilot a lot.
However, it introduced a very subtle bug where it used a loop counter `i` where it should have been indexing into an array like `arr[i]`, but it still typechecked and happened not to cause insane errors. Took a lot of time to track down later, and it's a bug I never would have written in myself.
For example you expect: x[i]
But copilot suggests: x[i+1]
…and you think, and you think, and then one of two things happens. Either you satisfy yourself that the answer is indeed x[i], or you accept x[i+1] and sheepishly thank Copilot for catching a bug today.
You could have copilot on in the background and never accept a suggestion, and it’s still a productivity enhancement.