while nearly no one is complaining about MS using github to train their copilot LLM,
Lots of people complained about that. I’ve only seen this single thread complaining about this.
Comment on China is attempting to mirror the entire GitHub over to their own servers, users report
raspberriesareyummy@lemmy.world 4 months ago
With the obligatory “fuck everyone who disregards open source licenses”, I am still slightly amused at this raising eyebrows while nearly no one is complaining about MS using github to train their copilot LLM, which will help circumvent licenses & copyrights by the bazillion.
JackbyDev@programming.dev 4 months ago
Kusimulkku@lemm.ee 4 months ago
nearly no one is complaining about MS using github to train their copilot LLM
What rock have you been living under??
Cosmicomical@lemmy.world 4 months ago
Came here to say this. As much as I don’t like china, there is really nothing to see (apart from the source, that’s for everybody to see).
mightyfoolish@lemmy.world 4 months ago
This could be illegal for git repos that do not have a open source license that allows mirroring or copying (BSD, Apache, Mit, GPL, etc.) Sometimes these repos are more “source available” and the source is only allowed to be read, not redistributed or modified. I would say that this is more of a matter for each software owner, not Microsoft.
But ultimately, this really isn’t as big of a deal as people are making.
Maggoty@lemmy.world 4 months ago
China is a sovereign entity. I’m pretty sure they can decide foreign licensing laws don’t apply there.
mightyfoolish@lemmy.world 4 months ago
China is a soverign state and they should make their own laws. However, China has promised repeatably that they will take IP concerns more strictly (trade deal with Trump in 2020 is one example of this promise). It seems of this moment they still use the World Intellectual Property Organization for inspiration for their IP laws. At one point, China did not acknowledge IP rights at all. Being consistent is good for business; especially when it comes to international business.
In 1980, China became a member of the World Intellectual Property Organization (WIPO). As of at least 2023, China’s view is that WIPO should be the primary international forum for IP rule-making. - Wikipedia
ILikeBoobies@lemmy.ca 4 months ago
Are you just trying to make a bad pro-China argument or have you never been online before?
Petter1@lemm.ee 4 months ago
I see it more as a good anti-Microsoft argument 🤷🏻♀️
ILikeBoobies@lemmy.ca 4 months ago
“Why does no one say murder is bad unless China is murdering”
Isn’t a good anti-murder argument
raspberriesareyummy@lemmy.world 4 months ago
“Why does no one say murder is bad unless China is murdering”
I can not fathom how you absolutely nailed the essence of my comment, yet misunderstood it (and - arguably - your own example) so fundamentally.
Let me try to help, once:
“Why do most people not complain about murder when Microsoft is doing it, but when China is doing it, the very justified outrage can be heard?”
Appoxo@lemmy.dbzer0.com 4 months ago
Not like MS couldn’t be sued.
It may be expensive but possible.
Unlike China. Good luck suing china (or the chinese government) as a whole. Maybe you’ll get out a domestic ban but I can hardly believe that they will care and probably will continue with their operation. But now it’s not on very legal grounds.kava@lemmy.world 4 months ago
If I look at a few implementations of an algorithm and then implement my own using those as inspiration, am I breaking copyright law and circumventing licenses?
sugar_in_your_tea@sh.itjust.works 4 months ago
That depends on how similar your resulting algorithm is to the sources you were “inspired” by. You’re probably fine if you’re not copying verbatim and your code just ends up looking similar because that’s how solutions are generally structured, but there absolutely are limits there.
If you’re trying to rewrite something into another license, you’ll need to be a lot more careful.
kava@lemmy.world 4 months ago
What’s the limit? This needs to be absolutely explicit and easy to understand because this is what LLMs are doing. They take hundreds of thousands of similar algorithms and they create an amalgamation of it.
When is it copying and when it is “inspiration”? What’s the line between learning and copying?
sugar_in_your_tea@sh.itjust.works 4 months ago
I disagree that it needs to be explicit. The current law is the fair use doctrine, which generally has more to do with the intended use than specific amounts of the text. The point is that humans should know where that limit is and when they’ve crossed it, with motive being a huge part of it.
I think machines and algorithms should have to abide by a much narrower understanding of “fair use” because they don’t have motive or the ability to Intuit when they’ve crossed the line. So scraping copyrighted works to produce an LLM should probably generally be illegal, imo.
That said, our current copyright system is busted and desperately needs reform. We should be limiting copyright to 14 years (as in the original copyright act of 1790), with an option to explicitly extend for another 14 years. That way LLMs can scrape comment published >28 years ago with no concerns, and most content produced >14 years (esp. forums and social media where copyright extension is incredibly unlikely). That would be reasonable IMO and sidestep most of the issues people have with LLMs.
raspberriesareyummy@lemmy.world 4 months ago
As I am a big proponent of open source, there is nothing wrong even with copying code - the point is that you should not be allowed to claim something as your own idea and definitely not to claim copyright on code that was “inspired” by someone else’s work. The easiest solution would be to forbid patents on software (and patents altogether) completely. The only purpose that FOSS licenses have is to prevent corporations from monetizing the work under the license.
kava@lemmy.world 4 months ago
Well let’s say there’s an algorithm to find length of longest palindrome with a set of letters. I look at 20 different implementations. Some people use hashmaps, some don’t. Some do it recursively, some don’t. Etc
I consider all of them and create my own. I decide to implement myself both recursive and hash map but also add certain novel elements.
Am I copying code? Am I breaking copyright?
As for forbidding patents on software, I agree entirely. Would be a net positive for the world. You should be able to inspect all software that runs on your computer. Of course that’s a bit idealistic and pipe-dreamy.
raspberriesareyummy@lemmy.world 4 months ago
again, I don’t have a problem with copying code - but I as a developer know whether I took enough of someone else’s algorithm so that I should mention the original authorship :) My only problem with circumventing licenses is when people put more restrictive licenses on plagiarized code.
And - I guess - in conclusion, if someone makes a license too free, so that putting a restrictive (commercial) license or patent on plagiarized / derived work, that is also something I don’t want to see.
sugar_in_your_tea@sh.itjust.works 4 months ago
I complain all the time. But that’s not the subject of this post…
PlexSheep@infosec.pub 4 months ago
Yeah exactly, fuck llms that don’t honor licenses