Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GitHub Copi­lot inves­ti­ga­tion

githubcopilotinvestigation.com

Post reply on HN
Print view View original

981–990 of 1001 posts

Re: GitHub Copi­lot inves­ti­ga­tion

#981

Post by sensanaty »

wilsonnb3 wrote: post #548
gernb wrote: post #364

Earlier quoted context omitted.

My 2 cents, 1. I think if you don't want your code re-used in CoPilot you should have that right 2. I think if CoPilot gets smart enough that it can read your open source code and then reproduce the algorithms without copying your code that should be fair use. It's the same thing a human would do. AFAIK CoPilot can not do that but I can certainly imagine it's not too many years away from that. 3. I think I would opt…

> 1. I think if you don't want your code re-used in CoPilot you should have that right People already have that right - all you have to do is not host your code on GitHub.

Someone could still take your code if it's hosted elsewhere and put it up on GitHub, at which point it gets sucked into the blackbox that is Copilot

Re: GitHub Copi­lot inves­ti­ga­tion

#982

Post by professorpiggos »

waihtis wrote: post #926
kurtreed wrote: post #925

To me the whole point of open source is selfless giving and sharing. You build something and release the source code in case it's useful for whatever purpose people might have: learning, understanding, contributing, forking, copying, etc. And companies might build on it, train models from it, use it internally, who knows. Great. Other companies can do the same and compete. So can other open source projects. For some…

Because a company of Microsoft's magnitude will build a walled garden around their ecosystem over time? Haven't we seen this effect in play like a million times?

If by walled garden you mean that their service is better than competitors, that isn't necessarily a bad thing. Codex does nothing to lead to a walled garden, it is just providing a useful service and could be the spark of more competition.

Re: GitHub Copi­lot inves­ti­ga­tion

#986

Post by amai »

tapia wrote: post #701

I think that Microsoft should train copilot with their own code (they own certainly enough lines of code after all). If they think that that would not be a fair use, then why should be a fair use to use somebody else's code?

Nobody would like to use Copilot if the quality of the code it produces would be like code from Microsoft. Garbage in = garbage out.

Re: GitHub Copi­lot inves­ti­ga­tion

#987

Post by woah »

jakelazaroff wrote: post #506
woah wrote: post #491

It would be sad if someone succeeded in shutting down CoPilot for this kind of copyright stuff. It is genuinely useful. I don't care that it reproduces copyrighted content. The only way you can get it to do that is to bait it with the function names of functions that have already been copy and pasted thousands of times onto GitHub without proper licenses. Luckily, someone will probably come out with a "renegade" vers…

> It would be sad if someone succeeded in shutting down CoPilot for this kind of copyright stuff. It is genuinely useful. I don't care that it reproduces copyrighted content. Sure — in the same way that hacking into a competitor's GitHub account and copying their private source code is "genuinely useful" to you. As the person benefitting from unlawfully using their source code, of course you wouldn't care that it rep…

> The only way you can get it to do that is to bait it with the function names of functions that have already been copy and pasted thousands of times onto GitHub without proper licenses

What? You left out the second line of the quote. It never reproduces copyrighted content for me because I'm not trying to bait it into doing that.

Re: GitHub Copi­lot inves­ti­ga­tion

#988

Post by woah »

Test0129 wrote: post #682
woah wrote: post #491

It would be sad if someone succeeded in shutting down CoPilot for this kind of copyright stuff. It is genuinely useful. I don't care that it reproduces copyrighted content. The only way you can get it to do that is to bait it with the function names of functions that have already been copy and pasted thousands of times onto GitHub without proper licenses. Luckily, someone will probably come out with a "renegade" vers…

> It would be sad if someone succeeded in shutting down CoPilot for this kind of copyright stuff. It enables the large scale theft of code. It completely ignores licenses. There are plenty of open source licenses that allow code use with proper attribution yet Copilot doesn't (and probably can't) figure a way to comply with all of them. Copilot, as the article suggests, is a marketing stunt. To me it's more than that…

> You can't just take 20 lines of completely stolen code, modify a few things, and call it your own.

That's exactly what I said in my comment. I wouldn't take 20 lines of code, since it wouldn't actually work. Even if it was able to spit out 20 lines of correct code, they would be tailored to my codebase, and not violating copyright.

The only time you see CoPilot violating copyright is when someone coaxes it into that, in a completely empty codebase with no context. The violation of copyright is not possible when it is used as intended.

Re: GitHub Copi­lot inves­ti­ga­tion

#989

Post by Narishma »

wilsonnb3 wrote: post #548
gernb wrote: post #364

Earlier quoted context omitted.

My 2 cents, 1. I think if you don't want your code re-used in CoPilot you should have that right 2. I think if CoPilot gets smart enough that it can read your open source code and then reproduce the algorithms without copying your code that should be fair use. It's the same thing a human would do. AFAIK CoPilot can not do that but I can certainly imagine it's not too many years away from that. 3. I think I would opt…

> 1. I think if you don't want your code re-used in CoPilot you should have that right People already have that right - all you have to do is not host your code on GitHub.

No, the only way is to make your code closed source.

Re: GitHub Copi­lot inves­ti­ga­tion

#990

Post by Narishma »

wilsonnb3 wrote: post #558
nmilo wrote: post #138

Good bye and good riddance. Even just the idea that GitHub should be allowed to train their proprietary AI on other people's work is insane. Much less distribute that AI in a paid package which lets you spit out other people's code verbatim. Anyone who supports open-source and the (ab)use of copyright law to create free works should be vehemently opposed to Copilot.

> Even just the idea that GitHub should be allowed to train their proprietary AI on other people's work is insane. You explicitly agree to this when you upload code to GitHub. FOSS folks shouldn’t have sold their soul to the proprietary devil but they did and now they have to deal with it.

How can you explicitly agree when anyone can upload your code to github?
Post reply on HN