"What can I do to reduce GitHub Copilot’s suggestion of code that matches public code?
We built a filter to help detect and suppress the rare instances where a GitHub Copilot suggestion contains code that matches public code on GitHub. You have the choice to turn that filter on or off during setup. With the filter on, GitHub Copilot checks code suggestions with its surrounding code for matches or near matches (ignoring whitespace) against public code on GitHub of about 150 characters. If there is a match, the suggestion will not be shown to you. We plan on continuing to evolve this approach and welcome feedback and comment."
Their approach of "matches or near matches (ignoring whitespace)" is clearly inadequate, and it's honestly insulting that they think this is enough. Even if Copilot just changed the case of a single letter, their filter wouldn't catch it.
>Notice that Copilot often gives code that verbatim matches opens source software, even when that filter is on.
I saw a few examples, but I don't see how that extrapolates to often. It's quite possible I've missed something in the article since I kinda skimmed it. :)
>and it's honestly insulting that they think this is enough.
They don't. - "We plan on continuing to evolve this approach and welcome feedback and comment."
Note that they opened both in the same vs code instance. And copilot uses other files in your vs code project as context to make predictions, so it could have reproduced this code without knowing it before.
I didn't say anything to dismiss or discount that some people just don't care or have a different view. I carefully qualified my comment to only address people who don't know, and they do exist in large numbers.
Or maybe they do know about it, and don't agree with you. Do you allow for such an option?
https://github.com/features/copilot
"What can I do to reduce GitHub Copilot’s suggestion of code that matches public code?
We built a filter to help detect and suppress the rare instances where a GitHub Copilot suggestion contains code that matches public code on GitHub. You have the choice to turn that filter on or off during setup. With the filter on, GitHub Copilot checks code suggestions with its surrounding code for matches or near matches (ignoring whitespace) against public code on GitHub of about 150 characters. If there is a match, the suggestion will not be shown to you. We plan on continuing to evolve this approach and welcome feedback and comment."