Doesn't really explain how co-pilot is stealing your community. I've used co-pilot and it works great until you are past boilerplate than it falls apart.
> Over time, this process will starve these communities. User attention and engagement will be shifted into the walled garden of Copilot and away from the open-source projects themselves
The author seems to be implying that since Copilot can reproduce the code of open source repository X in certain scenarios there'd be no reason for programmers to learn/use/engage with repository X. But this is silly. Maybe some open source repositories could be tab completed with a little prompting but people will presumably choose to add a dependency instead of tab completing the code of express or something.
It also doesn't make any sense. Copilot suggesting to me the signature of a function from some library is not actually the same as executing that library. That library still needs to be downloaded onto my computer to be executed. And who will write new features to a library if not for the people who are interested in that?
It strips the (mandatory, in a lot of cases) licence text. But the licence still applies.
(Or I guess more technically, the original authors copyright still applies, and the rights granted to use the work under the license as an exception to the strict limitation under copyright - do not apply...)
Agree, the argument seems pretty threadbare. Millions of programmers use open-source software every day and incorporate it into their own projects without ever engaging the authors of the code upon which they rely.
Perhaps the author means that there’s a possibility that the programmer to whom the code was suggested won’t necessarily know its provenance and how to engage the community from whence it came. If so, that’s a stronger argument, but I don’t know that it’s the best one they can make.
It's just basically a search engine, but the context is the rest of your code.
Writing software just auto-completing from CoPilot would be like trying to write a whole novel with the auto-predictive text on your phone. You could do it, but the results would be non-nonsensical and full of semantic errors. I don't think there's any real 'provenance' at play in either case.
This guy is a literal who, who is really over estimating the value of his open source contributions over a general development tool that can reduce the cognitive load of working in some hairy code bases.
I don't think it's CoPilot that's 'erasing' his, Racket, community.
"It's not a problem that it steals code because it doesn't work that well right now" is not a valid argument, don't pretend that AI doesn't advance on a daily basis.
Reminds me of people who defend AI art with "you can tell it apart from real art" yeah no, at the rate we're moving you won't be able to tell at all in a year or two.
I agree with your starting point, but art actually makes the questions clearer I think - having AI art "you can't tell apart" is possibly a bad thing for human producers of art (or not, obviously lots of artists don't make their art just to get paid, but if you do...), but seems like a great thing for the human consumers of art? Copiliot/tabnine etc. could certainly give you terrible suggestions, but my experience is they often help with boring scaffolding-stuff which isn't likely to be full of bad ideas, just time savings. Could they mess you up? Oh for sure. But mostly they just seem like an autocorrect that doesn't guess wrong every other time.
They don't really explain in a satisfactory way how Copilot is "stealing communities" even though they themselves complain that Microsoft hasn't provided "solid legal references".
Copilot is an AI stunt, an exploration, trying something new and exciting with very mixed and not-so-useful results.
This lawsuit, however, is just lawyers doing what they do for fun. I guess the retained Microsoft lawyers love it too. Glad to see lawyers having so much fun and profit. But we would all be better off with out so much lawyering, can't they do something more worthwhile?
You are willfully ignoring thst the article absolutely DID address this point!
The license and attribution are stripped from regurgitated copied code snippets from code projects.
If the people don’t known which project the code was taken from, how can they one day contribute to that codebase?
If the code projects on GitHub are not getting the people who use their code at least aware of the project, that project disappears.
Copilot is an interloper who doesn’t even tell you which project the code snippet was ripped off from!!
I’d guess it is basically: you’re making a mural people can look at for free, but a business selling canvas reproduction of it without attribution.
If I solve a problem and I say: Sure, use my code if you want, but be sure to contribute any improvement back to the community - I wouldn’t be happy seeing a tool spewing it out everywhere. And it’s not even free. In this case, microsoft is literally making money on the back of millions of programmers. And without approval.
The community argument is weak but it’s just content marketing, not a thesis. The goal of the article is to generate leads for potential members of a class action lawsuit.
It doesn't, it's just a very easy to relate to argument.
Generating snippets of code has nothing to do with a fully functional software package/product/service and an organic community around it.
One might argue that a community could be more easily formed thanks to co-pilot because it increases developer productivity and lower the effort to contribute so OS projects actually benefit from co-pilot. If this sounds far fetched, then probably the first case is also similar.
I'm more worried about the status of freedom in software, open source feels like a mirage to divert the attention away from the original issues from the FSF.
Great, I hope it is tried in court. It should be. But unfortunately I have not a big hope that the courts will come to understand the issue well enough.
Many courts, especially those in the Northern District of California (where a case would likely end up litigated), are very proficient and literate about software and copyright law. See Judge William Alsup’s cases if you want to see some examples that illustrate the court’s competence. And these judges frequently have technical consultants on staff to assist with technological issues.
Ok, well that sounds nice. I have little to no insight to how courts works in the states so I was talking about my experience of the court system at home :)
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
3. All advertising materials mentioning features or use of this software must display the following acknowledgement:
This product includes software developed by the organization.
4. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
5. Use of this source code for the research or training of machine learning models is permitted.
so what? You can put on a cowboy hat and larp as one, but that does not mean everyone else around you have to take it seriously. Same way with these made up licenses. If it's on the internet, it belongs to all. Or else keep it with yourself.
So, if you are not compliant with the posted license, you are in violation of copyright.
This is no different whether you are honkler or Microsoft (other than in how vigorously or not someone may enforce it).
If you really believe that anything posted to the internet ‘belongs to all’ then I don’t know what to tell you other than you live in a fantasy land where Oracle Corporation does not exist. We might all prefer it if things were that way, but they simply aren’t, and that’s just tough.
Only if you agree it is "redistributing" anything!
Small enough pieces of code can't be copyrighted. No one would support an argument that I violated copyright by using the code "else if {" from some GPL library.
So the question becomes what is the minimal unit of copyrightable code? What if you wrote a nice big function exactly (or almost exactly) the same way as someone else did? Whose copyright are you violating?
Does GitHub not have the right to view and train from your content when you agree to their Terms of Service and upload your code?
People are conflating their open source license with the one they give GitHub when making a GitHub account, but they are two entirely separate and parallel licenses. The former is for other people to use your code, the latter is for GitHub to host your code.
If you don't like it, you are free to host your code on your own servers.
And anyway, as noted the other day about AI, it is often funny to see people not care about (or even enjoy) AI in other fields that they don't work in, but when it comes for their own field, they are suddenly very worried. See programmers on HN who argue for Stable Diffusion but against Copilot, and vice versa with artists on Twitter. As I commented then, it's an act of cowardice to think our own profession should be immune from AI while we enjoy the fruits of AI in other fields [0]:
> Yes, many of us will turn into cowards when automation starts to touch our work, but that would not prove this sentiment incorrect - only that we're cowards.
>> Dude. What the hell kind of anti-life philosophy are you subscribing to that calls "being unhappy about people trying to automate an entire field of human behavior" being a "coward". Geez.
>>> Because automation is generally good, but making an exemption for specific cases of automation that personally inconvenience you is rooted is cowardice/selfishness. Similar to NIMBYism.
We should want AI. That we then try to use outdated models like copyright to enforce holding back human progress is a true shame. In my view, so what if GitHub uses people's code for training data, we are all getting a better product because of that.
Not all code on GitHub was uploaded by the copyright holder. The entire linux kernel is on GitHub and at least some of those copyright holders have never explicitly granted a license to GitHub beyond the GPL.
Code pushed to GitHub is quite often not pushed by actual copyright holders and there no way to distinguish between it even if there was clause like this in GitHub user agreement.
ToS isn't some all-powerful thing, first of all. A lot of it is unenforceable nonsense. And I'm not sure how it really works with OSS.
For instance, what if I self-host an OSS project but someone puts a mirror on GitHub? Or just uses GH as a remote for their fork? Does that random person accepting the ToS now mean GH has carte blanche to do whatever they want with that IP?
There are quite a few projects that didn't originate on Github. Some are mirrors of projects hosted elsewhere, some accept patches through other means, some include code that predates github. If get your linux kernel patch accepted by emailing it to the responsible maintainer, it will end up on https://github.com/torvalds/linux. But you never agreed to the Github ToS, all you did was agree to publish it under the GPLv2. Linus agreed to the Github ToS, but he can't give away rights he doesn't have, so he can't be giving Github any rights to your patches that go beyond the GPL.
We need the legal right to do things like host Your Content, publish it, and share it. You grant us and our legal successors the right to store, archive, parse, and display Your Content, and make incidental copies, as necessary to provide the Service, including improving the Service over time. This license includes the right to do things like copy it to our database and make backups; show it to you and other users; parse it into a search index or otherwise analyze it on our servers; share it with other users; and perform it, in case Your Content is something like music or video.
This license does not grant GitHub the right to sell Your Content. It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service, except that as part of the right to archive Your Content, GitHub may permit our partners to store and archive Your Content in public repositories in connection with the GitHub Arctic Code Vault and GitHub Archive Program.
That mentions everything: Parsing the content, showing it to/sharing it with other users, using it to improve and provide the service. GitHub and all of its features are "the service".
True, but it doesn't mention doing so without the attribution that might be required by the code's licence. If full attribution of where the suggestion was derived from was included¹ there would be not issue IMO², it is this matter that creates the grey area which these discussions result from.
--
[1] the practicality³ of this is a different, though related, discussion
[2] because the user is fully informed and can take responsibility for the decision to use the suggestion or not
[3] or impossibility – given the code could be added by someone who doesn't include that attribution/licence information for the system to be able to pass on even if it were designed to
The terms of service are completely independent of the code's license. The code could say "no one but I may use this", but by using GitHub you give them rights to do everything stated in the Terms of Service.
But if the terms of service says nothing that is in contravention of your licence choice when you agree to them, then the service does something that you consider to be in contravention of your licence choice, what you have is one party unilaterally changing the agreement. Of course the exact legal meaning of the terms and any perceived change in them could and will be debated long, hard, and potentially expensively…
I'll stick to self-hosting instead of using services like GH. Keeps things a little more simple in that regard.
The license agreement is irrelevant. Literally it does not come into play here. Github is not bound by the license; they are bound by the terms of service. The code is co-licensed: once however you declare it, once to Github independently.
Sharing with license intact. If GH is sharing with the license and attribution stripped, then just punting IP vetting to pilot users, it seems to exceed their rights.
There's nothing about "license intact" in those clauses. GitHub is able to do whatever it wants with the data; any users of the service do have to check on licenses, as they should with any source (including copying from Stack Overflow)
Cameras and paint brushes can easily make non-infringing works. Users of them can easily be trained how to avoid taking others work.
Copilot on the other hand basically defaults to infringing behavior. Users would have to go to great lengths to be sure they aren't infringing on others work.
Why do people refuse to have even a pre-high-school level of of understanding of licensing? By uploading your code to Github you are granting them their own license to the code under their terms. Your LICENSE file has absolutely nothing to do with it. Your LICENSE file can say "everyone but Github" and wouldn't matter one jot because that's not the license you licensed it to them under.
And if you didn't have the rights to grant the licenses to Github? Then you are in violation of the copyright holder's rights, not Github.
The only remotely plausible, yes-I-have-graduated-fifth-grade argument against Github is that they ought to and certainly do know that huge portions of their users are in fact granting them licenses without the necessary authority. That's an interesting argument we should be having, and instead we're having this inane screaming match by people who have no clue what they're talking about while some of us are sitting here going WTF is wrong with you?
I am discussing the license grant GH includes in its terms. And that doesn't appear to give them a blank check to do anything they want with code those users have uploaded. Certainly not sell it piecemeal.
IANAL, but it's pretty clear that GH explicitly says they will NOT distribute the code. I'm not sure what else you'd call offering to copy a section of code.
"It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service."
Throughout the license, the Content is treated as an indivisible unit, and it specifically refers to the forking functionality. Notice that forking...forks an entire repository, licenses included, etc. You can't fork a single file, and you can't fork a region of a file. GH provides that kind of forking.
Copilot is fine-grained forking.
No significant software company is going to permit copilot to be used and potentially poison their code base in unknown ways, now that this kind of copying is in the open and is clearly a significant danger.
Somebody like Black Duck is going to make a lot of money for trial attorneys by tracing how code was created and finding the "hits". That will be joined with log data indicating who used copilot, when they used it, and exactly what copilot presented as the "hit". This entire process will be performed recursively on the "hit", together with classic source analysis, to find out where something is really from.
The bigger companies are really, really serious about not copying outside code except under really strict conditions -- these conditions mostly look like "no you may not, unless you have one of these specific situations". It's no-by-default, even when it looks like it could be a yes.
Once someone uses copilot they now have code fragments outside GH and stripped of attribution and license. Looks like GH is trying to say these users need to do their own IP vetting. Which seems very impractical for anyone, even the creators of GH copilot.
> This license does not grant GitHub the right to sell Your Content.
This would, at a minimum, preclude charging for Copilot.
This is missing the point though. Microsoft claims their use of source code for Copilot is fair use. If they are correct about that, licenses don't matter, this EULA doesn't matter, etc. Everyone should be focusing on this claim, arguing about any other detail before that is decided is a waste of time.
> If you don't like it, you are free to host your code on your own servers.
From the article:
> “Dude, it’s cool. I took SFC’s advice and moved my code off GitHub.” So did I. Guess what? It doesn’t matter. By claiming that AI training is fair use, Microsoft is constructing a justification for training on public code anywhere on the internet, not just GitHub.
And:
> when it comes for their own field, they are suddenly very worried
From the article:
> First, the objection here is not to AI-assisted coding tools generally, but to Microsoft’s specific choices with Copilot. We can easily imagine a version of Copilot that’s friendlier to open-source developers—for instance, where participation is voluntary, or where coders are paid to contribute to the training corpus. Despite its professed love for open source, Microsoft chose none of these options. Second, if you find Copilot valuable, it’s largely because of the quality of the underlying open-source training data. As Copilot sucks the life from open-source projects, the proximate effect will be to make Copilot ever worse—a spiraling ouroboros of garbage code.
> As Copilot sucks the life from open-source projects, the proximate effect will be to make Copilot ever worse—a spiraling ouroboros of garbage code.
If I was writing this website, I would delete this sentence, because it is actually really idiotic.
1. This is an opinionated statement, except that there's nothing backing this opinion. This is fearmongering, that GitHub Copilot will get worse unless we sue Microsoft.
2. Lawsuits are not about concerns about a product's creator potentially damaging their own product. Lawyers suing Microsoft don't get a "we're protecting Microsoft from Microsoft's own bad decisions!"
The argument, completely unfounded, is that GitHub Copilot will undermine... GitHub Copilot. So, if you like GitHub Copilot, you should also be on board with suing Microsoft, so that we don't damage GitHub Copilot. What??? Good luck proving that line of argument in a court - you'd get laughed out of the room. Courts don't react well to hazy predictions about mayhem, from the suing lawyers, that have no historical facts to base them on.
How many millions of account were created before Co-pilot existed at all? It certainly wasn't in the ToS then except maybe in some extremely vague way.
> Does GitHub not have the right to view and train from your content when you agree to their Terms of Service and upload your code? [...] If you don't like it, you are free to host your code on your own servers.
I do exactly this, and it does all of bupkis to prevent someone from downloading my code from my gitea, and uploading it to GitHub. In fact, several people have.
Even if it does, it may not matter. For example, APIs are not copyrightable (see Google v Oracle), and if there is only one obvious efficient way to make something work, it does not follow that the user must be prohibited from using that way even if someone else did it first.
in countries where there is no fair use (most of the world outside the US) it seems quite likely copilot is willful, commercial scale copyright infringement
Fair use is unusually permissive in the US, but most countries have very complex copyright rules to allow e.g. a televised interview in a room with contemporary paintings, without getting permission from the copyright holders of those paintings. It'd certainly make for interesting cases.
> Microsoft characterizes the output of Copilot as a series of code "suggestions". Microsoft "does not claim any rights" in these suggestions. But neither does Microsoft make any guarantees about the correctness, security, or extenuating intellectual-property entanglements of the code so produced. Once you accept a Copilot suggestion, all that becomes your problem:
> "You are responsible for ensuring the security and quality of your code. We recommend you take the same precautions when using code generated by GitHub Copilot that you would when using any code you didn’t write yourself. These precautions include rigorous testing, intellectual property scanning, and tracking for security vulnerabilities."
I can't help but recall:
"Linux is a cancer that attaches itself in an intellectual property sense to everything it touches."
Fun story: That was my first employee town hall, in 2000. I was concerned for the fellow (and so very glad when he left, Satya has been so so so much better for the company and morale). It was definitely an... interesting introduction to the company.
At the time, I was doing Linux, OpenBSD and FreeBSD stuff in Bellingham. The reaction from the local and regional non-Microsoft community was really like "Holy shit what is going on down there?!"
It isn't so much a connection as an example of cognitive dissonance from the organisation.
On the one hand stating plainly that mixing in copy-left code and similar can be disastrously dangerous because it is a rampant virus. On the other hand not understanding why people think it might be a problem that their tool could encourage mixing in copy-left code.
Linux is open source and Ballmer is displaying Microsoft’s negative attitude towards open source that is demonstrated in the author’s arguments regarding copilot.
The point is not clear, but if I were to guess, it's that Github Copilot should come with a California Prop 65 warning, because it can give your code "cancer" (GPL-licensed snippets from sources like Linux codebas).
Microsoft released a product which gives you cancer the moment you use it.
According to the opinions about what inclusion of open source code into your projects does, as per the ex-CEO of the company. That seems a bit of a far fetched conclusion, but then, Ballmer did say it.
With "normal" code I can generally see (or figure out) who posted/published it and reach out for explicit permission. It's not uncommon for me to do this.
How is one supposed to do that for the generated stuff? Seems like an awefully hands-off attitude. As challenging as it is, they really ought to be qualifying the input samples of training code before ingesting.
I don't think it's right to characterize it as hands off after they had their hands all up in the generated code. It's just malfeasant. They've produced a tool that is fundamentally (legally) unsafe to use and said that's not their problem.
There are some techniques used mostly to detect when students copy paste code. I've seen some of the tools in that space and they have varying degrees of accuracy. MOSS is a common one[0].
There are some vendors in this space too (BlackDuck comes to mind) but they're $$$ so only within the scope of large corporations.
If anybody has any ideas relating to this type of analysis, I'd be excited to chat. I am working on a project[1] in this space for "Software Composition Analysis" which could potentially overlap with snippet detection for code like Co-Pilot. (We basically just have a big pipeline of analysis jobs that run on code and store the results. I need to update the docs!)
I think you need to explain that more. The problem (or at least one problem) being explored here is that by using any code from co-pilot, you are responsible for making sure the licensing is correct. You could unknowingly be using and modifying GPL-licensed code in your non-GPL project, which is a violation if you don't publish your modifications.
We're not talking about expanding copyright, just protecting the existing copyright systems from being trampled by microsoft.
For me it was the 'magazine' style with proper breaks and editing, formatted for a vertical screen but still reads great on my laptop. The effort in the content, links and emphasis make it feel like journalism I would normally get paywalled on.
With this site you see about 50-100 words on a large mobile screen. On HN you see 2-4x that.
Also, the section breaks and headers and boxes lack obvious rhyme or reason. It scans a tiny bit like a classy version of Time Cube. You keep getting hit with different font sizes and font styles and lines and ribbons and colors and you're not quite sure why.
Eh, I think the line height is too cramped and the hyphens make reading on the internet harder to read. Hyphens are good if you're trying to save ink or pages in a novel, but screen real estate is more than free on the internet.
I think the test for whether an AI is infringing or not should be:
Can this AI regurgitate the vast majority of the creative aspects of an original/novel piece of software with minimal prompting, to the point where the output code looks mostly and directly cloned to a reasonable person trained in the art?
_Maybe_ software is fundamentally different to other "creative works" which rely on copyright protection, but it's not immediately clear it is, and as far as I know it's certainly not a "special edge case" as defined in copyright law in general.
So "Can this AI regurgitate the vast majority of the creative aspects of an original/novel piece of software" is not the test that, for example, the music industry uses when determining if a sample is infringing. The test there is "is a sample, however small, identifiable as part of a copyright work by a reasonable person trained in the art?"
You can't own copyright in a composition of a single middle c note. But lawsuits have been won for copyright infringement of melodies of 2 bars (fewer than about 16 consecutinve notes). Men At Work lost a copyright case for the flute melody in Land Downunder which is the same as a 90 year old tune Kookaburra Sits In The Old Gunmtree https://www.claytonutz.com/knowledge/2010/february/men-at-wo...
Whether that's done by a flute player or an AI, really doesn't make any difference as far as copyright law sees things.
(Whether copyright law is a "good fit" for source code, and whether it makes sense to apply laws meant for books/literature/music/film to software is a different but very good question. I don't have much in the way of other ideas which take original author's efforts and potential rights to benefit from then though...)
FWIW, in this case I was trying to feel out what I think is a good fit for copyright in general. I think the same test could be applied for books, music, art, etc.
I really don't care if my code gets ingested and regurgitated by Copilot, but it seems rather a stretch to imagine that this is fair use, in part because it separates me from the legal protections afforded by the licenses I released my software under. In my ideal world, Copilot would be legally viable, and releasing my software without restriction wouldn't be risky.
As a long-time open source software developer, I have favored the 2-clause BSD and MIT licenses because they are the simplest licenses that provide me some liability protection. I would release code into the public domain if that didn't increase the likelihood of being sued, whether for liability, or for someone else claiming intellectual rights to code I actually wrote.
One issue I see with Copilot is that they get free access to all open-source data on GitHub, but using GitHub APIs to download the data yourself isn't possible (rate limiting). This is an unfair advantage. Copilot is not only making money off of open-source, they are making money off of open-source in a way others can't.
I would love to see a lawsuit which requires GitHub to provide their full Copilot dataset.
I'm gonna guess that Microsoft GitHub (tm) would shut you down pretty quickly if you tried to clone tens or hundreds of thousands of repos in a short window of time, b/c of course that's sketchy/abusive use of their infrastructure, right?
But of course if the data is already sitting in object storage inside your cloud environment and all you have to do is run some MapReduce jobs to get at it...
Hence: unfair, anticompetitive, intellectual-property-right-abusing behavior. Microsoft GitHub (tm) can prevent anyone else from running the kinds of analysis they do by simple "operational security", while running literally any kind of analysis, model training, etc. they want. Don't like it? But their commercial services and products so you can run Microsoft GitHub (tm) on your very own Microsoft Azure (tm) infrastructure, using Microsoft Visual Studio Code (tm) and Microsoft GitHub Codespaces (tm) so work on _your_ code privately.
Best of all, you can still still take advantage of the huge library of "free" code offered by Microsoft GitHub Copilot (tm) to ensure your private, proprietary codebase still has all of the advantages of Open Source Software, brought to you exclusively by the Microsoft GitHub Platform (tm).
I don't understand. Your favourite boba joint can email every one of their customers a coupon. That's "unfair" to the other boba joints without access to their mailing list too, right? You're just describing a regular old competitive advantage
> I'm gonna guess that Microsoft GitHub (tm) would shut you down pretty quickly if you tried to clone tens or hundreds of thousands of repos in a short window of time, b/c of course that's sketchy/abusive use of their infrastructure, right?
ArchiveTeam has a distributed Github archive project[0]. It's unclear what the status is right now. It seems like a worthwhile idea.
Actually they don’t. I’ve cloned thousands of repos before (tried to archive conda-forge org for a project).
I’ve also built many parallel repo downloaders for CI reasons. You can clone repos all day pretty much with little rate limiting. I haven’t pushed parallelism past 64 per host though
What do you mean "You're not supposed to"? Is there some law that forbids this? From my (potentially naive) POV this seems to be roughly equivalent to asking physics professors to stay away from mathematics since they are likely to have some relevant cross-domain expertise.
it only runs afowl of antitrust laws if it solidifies an unfair monopoly and/or uses market position as a monopoly holder to prevent fair competition from emerging.
most monopolies are completely legal. small town with only one gas station or one grocery store? 100% monopoly, 100% legal.
there are plenty of alternatives to github, paid, free, as a service, and self-hosted. GitHub has a large market share, but no monopoly.
This would, of course, need to be determined in court. Microsoft argued it had plenty of competition when the DOJ came for it over Windows and bundling. Like I said, this isn't unexplored territory for Microsoft. This is the company of Embrace, Extend, Extinguish and the Halloween documents.
This isn't a company with plenty of goodwill in its sails launching a hip new product. They're not Do No Evil era Google or Apple riding high on the iPod. We can't just pretend there isn't a history.
you're making a big assumption here, that Microsoft did not learn from their mistakes.
I posit that they have indeed learned from their mistakes.
1) They train Copilot with repos hosted on github.com. 2) Users who upload code to github.com grant GitHub an explicit license[0] granting GitHub the right to show that code to others. 3) GitHub do not specify which technologies or techniques they may use to show this code to others, meaning they may use any technique they like.
Microsoft have learned. And they have covered their collective asses. Users who host code on github.com agreed to these terms.
Users who don't like their code showing up in Copilot should not be hosting their code on GitHub, because they agreed to have their code delivered to others when they signed up.
> This license does not grant GitHub the right to sell Your Content. It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service[...]
I imagine the crux of this case will be what constitutes "the Service", and whether that includes Copilot. Also whether licensing Copilot counts as selling Your Content.
Microsoft got done because they were threatening vendors with punitive action if they didn't bundle Windows and IE by default.
Which is to say: Microsoft was taking specific action which wasn't a natural consequence of their software, but was being actively enforced to keep out competition. Vendors didn't organically discover consumers weren't interested in getting a PC without Windows and IE, they were prevented from even offering the option lest they be completely denied the ability to offer that at all.
maybe, but it is only an illegal monopoly if they are using their monopoly position to prevent others from entering the market.
they are not threatening to block access to github.com for all Comcast users if Comcast chooses not to block access to gitlab.com, for example. That would be an illegal activity for a monopoly. simply existing as a monopoly is not itself illegal.
"Github likely does have what would be considered a monopoly marketshare"
Citation needed. I'm interested to know of examples of how they have stifled competitors. AFAIK Gitlab came of age well after Github had already grown super-large. If Github could have killed Gitlab, wouldn't they have done so?
That is also excluding Azure, Bitbucket, AWS and the plethora of other git repository hosting companies and services. AFAIK, gitlab has more payed (AKA, private or premium) customers than does Github (though I lack a citation there, but that was my understanding when researching this a couple years ago on where to host a private companies code. My take-away then was that Gitlab is actually more popular amongst private companies and Github is more popular for open source projects).
What would be considered Monopoly marketshare? I would suggest that Facebook, Twitter and Amazon have monopolies. If you get shut down as an Amazon seller, that can be 90% of your revenue. Given so many people can leave, and have left Github voluntarily, is implicit evidence that there are very decent alternatives (if not, there would be no alternative and you would be forced to stay with Github for lack of alternatives. That is not the case though, it's easy to just go over to Gitlab. From my perspective, I have no idea how Github can kill Gitlab, I'm curious if there is a vector where Github could use it's network effects to diminish Gitlab, as a specific example. So, how could Github do that?)
<<<The Sherman Act outlaws "every contract, combination, or conspiracy in restraint of trade," and any "monopolization, attempted monopolization, or conspiracy or combination to monopolize." >>>
I know lots of people here don't like it, but it is the law and that was the question; "this" in parent clearly meant "use dominance in one market to gain dominance in another" in grandparent, regardless of whether that's actually the central issue here or not.
You might be confused but that's literally what you do as a business. You leverage your domain area expertise to expand into new areas of business.
For instance, Apple already knew about the portable hardware market and extended their reach into the portable music market via iPod. It used the iPod to reach the music marketplace via iTunes. Used its market dominance to create iPhone and the rest is history.
Maybe that's a personal maxim on your part but there's no law that says if you have dominance (what does that even mean?) in one market you cannot enter another. Think about what you're saying. We'd have no multi-product company if that were the case. What does "dominance" or "market" mean anyway?
That's what you do when you are not at-risk of being a monopoly in a sector. Apple could do these things back then because they were still a relatively small company in the tech sector. But now a days things do start to get more complicated.
As you point out, establishing what dominance is in a market is a tricky thing and it is why governments will investigate any potential signs of it. Now, personally I don't think this is really an issue for Github, or microsoft (the parent company) but let's not pretend market dominance and abuse of dominant position are not a real things.
TLDR: If you have a monopoly in a market, you can't use your position to get a heads up in another distinct market by bundling products together. Ex: Microsoft can't use their monopoly position in the OS market to get an advantage in browsers by bundling IE with Windows.
It's my understanding that this doesn't apply if you don't have a monopoly, and it also doesn't apply unless you're actually bundling the sale of multiple things together.
>You might be confused but that's literally what you do as a business. You leverage your domain area expertise to expand into new areas of business.
>or instance, Apple already knew about the portable hardware market and extended their reach into the portable music market via iPod. It used the iPod to reach the music marketplace via iTunes. Used its market dominance to create iPhone and the rest is history.
Look at what you're saying here. Apple had domain experience in hardware and launched a new product (good). Then it used the dominance of that product to muscle into an entirely different market (bad). And the combination of hardware and market has led to Apple being able to extract their tax on half the music market, or whatever they have.
This is exactly what we don't want and why anti-trust laws exist.
In the same vein that Google are not a monopoly in search, Microsoft are not a monopoly in consumer and business OSes, Chrome is monopoly in browsers, etc.
Just because there is some existing competition which has a few percent marketshare and technically it's not a monopoly doesn't materially change anything besides a pro forma excuse. Which is why Google have been propping up Mozilla, they want the excuse "but technically there's another browser". However for consumers and the market it doesn't matter that technically there's an option that practically nobody uses.
"git hosting" is too narrow a market. Code repository may be someone interested in, there Github won't have a practical dominance. Most probably by nature of complete Github, it will come under "software development tools"
There's a sizable fully adult cohort in tech who are young enough to, at best, have vague memories of Microsoft's antitrust troubles and not get why this is such a big deal. Tech history education could be better.
Yes. But the people who agree to let GitHub host their code are doing so with the expectation that it's "open source", i.e. freely accessible.
If we define "open source" as "you can't necessarily use this to train an AI", then Copilot itself is illegal because it's using code without permission.
If we define "open source" as "you can use this to train an AI", then Copilot it legal, but GitHub may be illegally misrepresenting itself as a host for "open source", as the policy it hosts code under isn't truly open-source.
If we define "open source" as "you can't necessarily use this to train an AI" but then GitHub's policy explicitly states "by using us as a hosting provider, you give us permission to use your code to train AI" then they are in the clear. But I doubt they have that clause or at least had it when Copilot was first revealed.
Not a supporter of Copilot, but I think it's pretty easy to access the same data through BigQuery:
>The Google BigQuery Public Datasets program now offers a full snapshot of the content of more than 2.8 million open source GitHub repositories in BigQuery. Thanks to our new collaboration with GitHub, you'll have access to analyze the source code of almost 2 billion files with a simple (or complex) SQL query.
There is a distinction between being able to access the source code, and a tool giving it to you without any context of the underlying license it is governed by.
GP was saying that GitHub has an unfair advantage in that they have instant access to all GitHub code, whereas everyone else is rate limited.
I'm pointing out that this limitation is not meaningful because everyone can access all GitHub hosted source code through BigQuery, where they won't be rate limited.
Don't forget that GitHub is a closed-source proprietary commercial service.
Their freemium product is useful to many open-source projects and communities, but you do not have any more rights to use Microsoft's GitHub than you have to Microsoft's Windows.
> using GitHub APIs to download the data yourself isn't possible
Is it? Data storage would be prohibitive, but I can see ways to download the entirety of Github in a few weeks/months (assuming my size estimate is accurate).
It's probably the reverse - the data can probably fit in a few commercially available hard drives, the API calls to discover and download all repository without running afoul of the rate limiters and whatever other anti-crawling strategies can take years if you're running it single threaded (a single IP can effectively be considered single threaded).
There are publicly published datasets, updated regularly, and you can stream all the events happening on Github (which is how secret leaks happen) so you can stay abreast of new repositories.
Anyone can start their own GitHub competitor and do whatever they want with the source code that ends up on it. GitHub pays the bills and lets us freely upload whatever we want to their service, so it seems a bit entitled to complain about what features or data they provide or don't provide.
You've done nothing to refute my point and so it still stands.
I pay GitHub / Microsoft to host my code, and that's all I expect them to do with it, host it, as securely as possible. It sounds like Microsoft are doing more than this so what's your actual deal...if you have one?
I did, what did I miss? I just re-read it and it's full of statements which are totally inline with what I expected:
GitHub considers the contents of private repositories to be confidential to you. GitHub will protect the contents of private repositories from unauthorized use, access, or disclosure in the same manner that we would use to protect our own confidential information of a similar nature and in no event with less than a reasonable degree of care.
FWIW, not all of us "happily pirate movies and music".
I want there to be more good music and movies. I want to support artists who create entertainment I enjoy. I go out of my way to buy physical copies of music from artists, wherever possible from the merch table at their shows or from their own websites. I pay to go see movies on the big screen (partly because I like the big screen cinema experience, but also because I understand "opening week revenue" is a key performance indicator for the success of a movie).
I thing copyright is old, outdated, and probably not really fit for purpose for forms of creative work invented in the last 50 years. But I also thing creative workers need to get paid for their effort (juist the same as software developers), and absent a FAANG-style set for companies employing teams of songwriters, musicians, authors, and the like - on FAANG-style salaries, copyright seems to be the option that is working (however badly).
I'll join your "abolish all copyright" crusade as soon as there's an alternative that at least likely to possibly work as well (or better) than the system copyright allows. Just abolishing copyright and erasing the publishing/music/movie/art industries without a transition plan isn't a thing I can support. (At least a transition the artists/editors/producers/writers/etc. I'll admit there's a large chunk of management and legal in the fairly abusive parts of the music industry I wouldn't shed a tear if they all became homeless and destitute overnight...)
This is a bit off-topic, but I wonder if there are people/teams right now creating git repos, doing the source code equivalent of "SEO" on it, and embedding backdoors in stupidly overoptimized for the training process code?
I wonder when we'll hear about the first big hack that gets traced back to production code pushed live after CoPilot "suggested" eval(base64decode({webshell}))
The less overt version of that is to figure out what mistakes copilot already makes (either things that are common in tutorials but not good in production, or things that are outdated, like hashing passwords with md5), and then systematically looking for software that includes such copilot suggestions.
Is there a technique to scan for software that includes copilot suggestions? Or is this just theoretical? Sounds impossible given MS/GH's monopoly on access to the model input data.
Probably not, but I can imagine something similar to hijacking a popular library and publishing a new version that opens a certain port and waits for instructions. All a malicious actor needs to do is increase the amount of exposed servers to be caught while they later scan the internet for anyone with that port open.
What is easier is to probably sneak in a new dependency that points to a malicious fork. Something like an XML to JSON library that does do the thing that is advertised but also additional things.
It seems like there's no good license that places absolutely no restrictions or requirements on people using your code (such as attribution and respecting patent rights) worldwide.
I want my code to be used the way people treated text in the old days. There's texts that have been re-written, added to and edited by thousands of people over the centuries and yet they don't come with thousands of pages of attribution notices because why would they?
You don't need to license it, you can just publish it with a declaration that as the author, you are releasing your work in to the public domain. However, in terms of licensing I believe MIT is the most permissive.
Okay, so just take "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software" out of MIT and call it the Do Whatever You Want Licence. You're not obliged as an author / copyright holder to impose restrictions on people using your work if you don't want to.
Whether Copilot is breaching MIT depends on what constitutes a substantial portion, which I am not qualified to rule on.
I do have to wonder if Copilot will last. It's going to become a legal minefield and I can't imagine for a second that Micrsoft will want to be in the crosshair for another antitrust case.
Yes because it turned out so badly the last time. Microsoft went from being one of the three most valuable companies in the US in 2000 to being one of the three most valuable companies in 2022.
Also back then, Microsoft had 90%+ share of the PC operating system market and was bundling IE in its operating system. I’m glad the DOJ forced MS to change its ways.
IMO, It will last. It's impossible to roll it back.
IMO, if the lawsuit goes to a point where it's likely to be won by copyright owners, Open AI could do the following:
- Use less sensitive code from big corps they have partnership with for training. I bet MS and other have plenty of such code.
- Buy training rights from copyright owners of OSS projects. Many of them have SLAs which allow the owner do much more than the license allows.
- Buy rights to train code, and collect generated code with Copilot from a large number of smaller software companies, likely with exclusions for some sensitive parts. MS has a lot of leverage here (discounts, partnerships, etc).
The example given is "sparse matrix transpose in the style of Tim Davis", but someone who wanted something with such specificity would be able to just take it from Github anyway, perhaps with a little more searching.
And would therefore have to follow the license of the code they took it from. That's exactly the point. Copilot is reproducing the same code but without the license.
Please look up the license in the source code that Tim Davis points too, it just mentions it's LGPL, but doesn't include the full license text. And none of the C code mentions a license or Tim Davis.
And if you dig further, the whole repo is mixed with BSD and LGPL "licensed" packages together. It's probably best that CoPilot does not suggest from code that does not have an explicit license stated.
I think originally Tim Davis was complaining about the non public sources for suggestions which Github CoPilot ignored.
A simple search on github reveals that those functions were reposted verbatim thousands of times, most people just copy and paste snippets of code they find useful, ignoring licenses. This highlights how all the power a license promises to hold is completely fictional. Any "in the style of Tim Davis" modifier only shows some kind of unwarranted self-importance complex on the part of the guy, thinking his style is widely known and distinctive (it's not). It's not the job of Copilot, the team that builds it, or the programmers that use it, to determine where the functions that were reposted thousands of times under all kinds of licenses originated.
This is the same case as with copyrighted photos in newspapers, a paper prints a photo somebody allowed them to use, but then it turns out that person did not have the right to use it in the first place. Did not stop newspapers from printing photos at all.
It does not show an unwarranted sense of self-importance on the part of Tim Davis.
Whether or not his style his widely known, his code is VERY widely used. Just look up SuiteSparse and try to find all of the downstream uses of it. It is one of the most---if not the most---ubiquitously used set of sparse linear algebra libraries. If you do anything with numerical linear algebra, there's a good chance you at least know what SuiteSparse is, and possibly also know who Tim Davis is.
The bigger issue here is the effect this has on research. Tim Davis not only programmed this library, he did the basic research leading to many of the algorithms in SuiteSparse. He went ahead and released SuiteSparse open source, probably thinking that it would be a good deal for him, provided that its use was properly attributed. Provide a public service in exchange for attribution. This is a reasonable way to get support as an academic. Clearly he has had a large number of industrial collaborations which likely have provided him with a significant amount of funding over the years.
Speaking for myself, if Microsoft has no compunction against behaving this way, I can no longer see the point in publicly releasing research code that I develop using an open source model. Microsoft is clearly telegraphing that they don't give a f** about licensing, although whether that holds if they are litigated against remains to be seen. I think there's an excellent chance many other researchers feel the same way. If you think openness and reproducibility in science is important, this is a problem.
100% correct takes in the piece, this is just ridiculous
>"Tim Davis gave numerous examples of large chunks of his code being copied verbatim by Copilot, including when he prompted Copilot with the comment / sparse matrix transpose in the style of Tim Davis /."
Copilot regurgitates code and blatantly violates licenses, not even sure what there is to argue about. Not only does it seem straight up illegal and sideline open source communities, I think the next logical step of this is that people who want to avoid having their work vacuumed up and their rights violated simply to move to proprietary software, which would be a huge disaster for open source.
There are lots of comments arguing for or against Copilot on a value judgment, and having an opinion on it being ethical or legal, etc isn't going to be the same for everyone. But I think regardless of where you stand, there should be some sort of legal ruling to clarify the gray areas that Butterick breaks down.
Agreed, but I also hate how so much of our substantive law basically has to be created by the courts because (a) many of our legislatures, especially at the federal level, have become more and more non-functional, and (b) IMO legislatures are especially bad at implementing technical legislation.
I think there is a good, fundamental legal/societal question of how copyright should apply to AI output. I just don't think our existing copyright structures handle this question well.
Note there is currently a very important case before the SCOTUS that is related to this issue, [1] where the original photographer of a Prince photo is suing Andy Warhol's estate for copyright infringement. The fundamental question is whether the Warhol series of painting are "transformative" enough of the original photo. While there are always gray lines on what "transformative" means, if there is any chance that Warhol's painting are legal and not infringing, I don't see how Copilot could be in the wrong. Copilot's output, even if it contains a substantial amount of the original source, appears to me much more "transformative" than the Warhol paintings are compared to the original photo.
That's how common law works, it's not disgusting, (unless perhaps you're an overzealous adherent of civil law) nor is it ex post facto. Legislation is produced, (claimed) grey areas are challenged in court, if the outcomes appear unfair then legislators (should) update the law.
Badly written law and poor legislators are a problem in any system.
And if that's the case, the legal ruling shouldn't stop at code. It should encompass any images or text that aren't in the public domain, don't have copyright owned by the training entity, or otherwise permit this use. Which probably throws a massive spanner into a lot of machine learning. Which may well be fine but would probably be a big setback for ML generally.
Any higher court ruling might well draw some lines between different domains, but be clear that a ruling against GitHub would almost certainly be a ruling for copyright maximization and against fair use in other respects. So be careful what you wish for.
Butterick sneakily asserts over and over that Copilot is simply retrieving code from Github ("Copilot's whizzy code-retrieval methods", "Copilot is merely a convenient alternative interface to a large corpus of open-source code", "our work is stashed in a big code library in the sky called Copilot"). This verbiage seems specifically chosen to present a misleading picture of what Copilot is and does.
Copilot is a set of trained weight values in a matrix. There is no source code stored in that matrix. The fact that someone can prompt Copilot with specifically chosen text to generate a short sequence of code that matches a corresponding segment of code used to train the model does not mean that it is somehow "just retrieving" that snippet. It is _generating_ that code, guided by the weight matrix, via pattern-matching based on the chosen textual prompt and surrounding context.
That distinction is significant because one of the primary defenses against copyright infringement in US law is if the derived work is transformative. Copilot is a work derived in part from Github code, but it has unique capabilities far beyond returning short snippets of input code, and the work itself is clearly an extensive transformation of the input data.
This is without even considering whether concrete _outputs_ of the model that happen to match code in a repository used to train it are themselves protected via copyright or not, which is another issue entirely (and not as cut and dried as many folks on here seem to think).
Correct. He's written a great opening argument, as long as you're the sort of person who likes speeches. To me it was full of tricks to prime the reader into accepting his premises as axiomatic, from nuanced rhetoric to pull quotes with attractive color gradients. In my view his actual motivation is the typical 30% cut of any class action settlement that goes to the lawyers, and he sees a lucrative opportunity to combine two skillsets.
Open source? They used everything on github with no regard for license, which would have included plenty of code under conventional copyright. Microsoft is now profiting from that code.
New tech creates winners and losers, and losers inevitably complain. See looms, VHS, Napster, etc. The more of this complaining I see, the more it falls flat. The only interesting thing is which side different communities end up being on.
To be fair, record companies were not in the least bit sympathetic. Open source contributors are easier to identify with, though imo it doesn't actually make their concerns more valid
He's saying you're a luddite for not wanting Copilot to steal all your code and use it for furthering the grand vision of AI generated code (which supposedly represents the forward march of progress)
Copilot exists publicly, which also means some copilot-lite thing trained on a smaller subset of repos probably exists privately in many different places. It may not be as good today, but these private instances will improve over time. Since the demand for a copilot-like service exists, eventually a large VC-funded public instance will show up.
In that lens, it is more sensible on the individual level to prepare for a world where copilot thrives than to put all of your eggs in the "ban copilot" basket.
Copilot wouldn't be shut down or neutered because "a few vocal people" protested against it.
It would be because it's illegal and violates the licenses, desires, and intentions of the thousands of workers who wrote the code in its corpus.
You act like Microsoft is trying to do a public service and people are angry about it. The reality is that they're taking billions of hours of work and using it to build a product that only they control.
If they re-released Copilot as FOSS, a lot of the valid criticisms would evaporate.
> It would be because it's illegal and violates the licenses, desires, and intentions of the thousands of workers who wrote the code in its corpus.
I wonder how many people on HN would be on the side of the creators if we were talking about content created by Walt Disney and whether pirating was ethical?
That many people who are criticizing Microsoft for using open source code and claiming “fair use” and saying it’s not fair to creators are the same ones that say pirating digital content is harmless.
Because issues aren't automatically symmetrical, and impact of actions isn't equal in both ways, especially when there's power difference between the sides of the conflict, e.g. it's not hypocritical to punish a bully punching their victim, but not the victim for punching the bully back.
Disney used its power to distort copyright laws in a self-serving manner. As an individual you don't have equal power to oppose them (you were supposed to have in a democracy, but lobbying is legal and corporations are people).
Disney is a huge corporation that won't even notice if you pirate a movie, which you may not even have been able to pay for anyway, because of their region-locked twisted maze of distribution and DRM.
OTOH you may be screwed if you're a creator making a living from your work, and a big corp can just take it without paying, launder it through "… in the style of $YOURNAME" query, and say they own it now, because unlike your copyright, their Terms Of Service apply.
Even people who think copyright shouldn't exist may rely on using copyright — against itself. You can't unilaterally say "I don't believe in copyright", because the law doesn't care, but if you license something as copyleft, then the law does care about your anti-copyright license.
There’s an assumption that hackers are automatically anarcho-communist, and unconditionally, wholeheartedly, vocally supportive of content piracy, which is not correct.
I don’t think anyone at all is here arguing that an AI trained on a massive corpus of movies that outputs snippets of film based on a prompt would be illegal. Such a thing would literally be the same as Midjourney with images. The fact that you can likely coax any AI to output snippets close to some of the source material is not likely to really matter and would be as if you recreated a copyrighted work using any other tools.
There is a certain amount of irony that a likely possible outcome will be the strengthening of copyright laws and enforcement, at which point people will realize they won the battle but lost the war
But isn't it copyright upon which open source is founded? The license is based on the right of the owner of the copyright to authorize someone else to create a derivative work?
Without copyright, it would be perfectly legitimate and legal for someone to not follow a license (because it would bear no legal weight because of the lack of copyright).
I would argue that open source is best served by strong copyright protections that allow the people created the software to make sure that further changes to it are released back to the community. Weakening copyright law means that it is that much easier for big companies to co-opt some software and not need to release the changes back.
That's a bit of an exaggeration. There have only been two copyright term extensions since Mickey Mouse was created, and only one of those can even remotely be attributed to Disney lobbying.
Even for the term extension that is attributable to Disney, you can also attribute it to Germany and the formation of the EU. "Mickey Mouse Protection Act" is funny, yes, but also an oversimplification.
This is also why I fully expect Steamboat Willie to fall out of copyright protection in January 2024 - right on schedule. There's a few countries that have supra-EU copyright terms, but none of them are dealmakers. Nobody is demanding we match Mexico's life+100 terms, for example.
I am 100% on the side of content creators. Regardless of who they are .
The courts tend to take a dim view of theft. Which is what this is.
The article clearly lays out that multiple requests for sound legal basis have gone unanswered . It simply doesn’t exist and Microsoft is operating on a forgiveness vs permission model.
Licensing is 100% about permissions. Clear and explicit enumeration of the permissions (or lack thereof ) for a work.
This class action lawsuit should surprise nobody. It’s a class that is sick and tired of being exploited.
Do not take my work that I contributed with explicit permissions and use it in a way I didn’t grant permission for. Full stop. It isn’t complicated.
so much animosity over rights you gave GitHub when you put your code there. "Theft" gimme a break. you license your code to GitHub so they can show it to others. This is separate to the stated license in your code. Nowhere in that terms of service document are the means that the code is shown to users specified.
Give me a break. Did you even read the license? You read that it is restricted to "the service". That service cannot be copilot, because it didn't even exist when most people agreed to the license. Moreover the license explicitly states they cannot use the code to distribute in any other way or sell it. So if anything they are violating their own agreement.
Also MS themselves don't even claim that training is covered by their terms of service, they claim it is fair use.
? Did you read what I wrote, MS doesn't even claim copilot is covered by their terms.
They claim it is covered by fair use (some people also claimed there is code not hosted by GitHub in copilot, which would further confirm that they believe they are covered by something else)
Moreover terms have been largely unchanged for years AFAIK. If someone agreed to the license years ago, they can't have agreed to copilot use. Also copilot is not a service on their website, it is a separate service and they charge for it, also contradicting the terms.
If you think I'm mistaken I will gladly reevaluate what I said if you could kindly restate your position, by answering a few questions
What does separate service mean? What would Copilot look like if it was not separate?
Elaborate on "not a service on their website", as it is available and listed as a feature "Github Copilot" on their website.
Is the contradiction related to payment for the service, or just because you think it is separate?
Since I thought you were arguing that Githubs own terms prevent them from using the public repositories in Copilot, this is what I argued against.
If you think fair use is involved, then that's the end of the line. If MS claims fair use, then until a court says otherwise, it is. Anyone who thinks their copyright is being violated can get an injunction tomorrow.
they claim the training of their model is covered by fair use, but they did not say that was the justification they were using. They don't need to claim fair use.
It's pretty clear from the Terms of Use that they can use code hosted on github.com to provide any service they like, so long as it is a GitHub service. They don't need fair use, they already have the rights to do what they are doing.
If my code is licensed under terms , that’s the permission .
If you use my licensed work then yes, you do need to follow the terms of the license .
The issue of license / contract / copyright is messy. It doesn’t ever seem (in the USA anyway ) to be definitively answered / “solved.”
I chose AGPL v3 only on purpose.
Co pilot and users thereof (so now two levels removed ) utilizing code in whatever work are stealing my work (unless it’s AGPL v3 licensed ). The adding of intermediaries (and the most likely unknown and with no way to know) infringing is going to be very difficult to mitigate. It’s like truly unknowingly buying stolen property,
If I use it under fair use, there is nothing you can do about it.
The fourth factor of fair use is the effect on the value of your work. If I'm not affecting the works value because there is no market for it, because it has no commercial value, you are going to have a very hard time defeating this argument in court.
> If they re-released Copilot as FOSS, a lot of the valid criticisms would evaporate.
That changes nothing at all. A FOSS license is, in a legal sense, no different then a proprietary one. If it’s illegal it’s illegal regardless of whether it’s FOSS.
If Copilot is deemed a derivative work of its inputs, it would resolve the issue that it does not comply with the AGPL because it currently does not provide source code to its users under an AGPL-compliant license.
Why should Copilot engineers, and the company that invests in it, not be rewarded for their incredible product and SaaS offering they spend resources on providing?
That's not theft. Was the original code deleted? No. Then it's copying. And not even that, is the model replicating the training set like Google search? No. Then it's some kind of derivative work. And especially for Github it's OK because user agreements allow MS to do it.
Is "imagining" the same with "copying"? Does copy-right cover learning-right? Can learning and practicing be restricted by the authors? Can visual styles, algorithms and facts be copyrighted? I say no to all of them.
Larceny and copyright infringement are two distinct legal categories for two distinct kinds of crimes. One involves a taking of property and the other involves the violation of a government granted monopoly
Human's aren't even allowed to do this. It has to be done clean room - this is not learning, it's copying and has been proven so.
The only implementation that would be allowed is if they had to describe the code via a specification and Codepilot was able to generate it from scratch. It does not do that - it just reorganizes stuff it's already seen, which is a copyleft violation in humans, therefore it should be for a machine created by humans.
That's for patents not for copyrights. All you need to do is make it a little different, generate until satisfied for patent safe code. Another model can also do the patent checking.
I've heard multiple stories from people who can't even look at the Linux source before they do something in their company's kernel because the act of just looking at it compromises them legally. So which is it?
Clean room implementation only happens for very special codes, like codecs or efficient matrix multmul that took billions of trials to develop with AI. Not for 2-3 lines of code snippets that do one simple thing and are already covered online in hundreds of places.
One of the canonical examples of software reverse engineering is Phoenix Technologies producing a compatible BIOS for IBM PC. They did exactly what OP described - had one person look at (public but copyrighted) IBM source and produce an extremely detailed design document, then another team went and wrote a new BIOS from scratch following that document. The issue at hand was copyright, not patents.
It’s not about “learning” at all, copilot spits out copyrighted licensed code verbatim directly copied from the source to your project in violation of specific and multitudinous repositories.
They are making rips of other peoples stuff, selling the contents of peoples “books/movies/songs” sans author attribution or album credits etc… to put it in terms you may be familiar with
Vinegar and salt on open wounds. Bad
Is there any nuance for scope?
Software I've licensed as GPL, I'm concerned about the working software being re-used and re-licensed for something commercial. For a given method out of tens-of-thousands, it's very not-germaine to the overall software to the point that I don't see it is a really relevant (but that is my opinion). Though, if someone likes the opening sentence of the encylopedia (or some other giant work), and an AI says, "this is a good opening sentence - does that really make for "ripping" off? Isn't the covered work the larger contents of the encylopedia, rather than an arbitrarily well written opening sentence? Isn't the big part of the work the ensemble?
I'm starting to wonder about these arguments, and whether we've gone into bad faith and hyperbole territory here. Are algorithms subject to copyright? Is it the case that if a GPL work uses a well known algorithm, that GPL work cannot be used as a reference? (Given that algorithms have very limited forms they can take, using an algorithm as a reference is really just copying it. Even translating pseudo-code to code, it's still the same thing).
Can you explain to me how something like using Eulers formula to solve a math problem would not be copyright infringement? A GPL project might use that formula somewhere, but then using that would be a copyright violation?
How about HTML source code, does putting a 'copyright' notice on the webpage make it invalid to then use any of the javascript, even if it has nothing special to do with the domain of the website?
"selling the contents of peoples “books/movies/songs”
Going to this analogy, I don't know if it is really the contents, but more like the first sentence, or even the first few words of that sentence rather than any recognizable subset of that work.
Like, if I have an app that does a spreadsheet, I don't care if you take an implementation of quick sort from my code as reference, but I do care if you use the same and main features of the spreadsheet app that I made.
They built it using some publicly available resources. These resources are available conditionally, subject to licenses (such ad GPL). Which is fine.
The problem is that their product sometimes produces verbatim copies of licensed works, without attaching licensing information. This not only goes against the licenses under which the original authors made these works available. It can also put the product's users in danger of anything from bad publicity to a copyright lawsuit.
CoPilot is a very interesting research project. It's not yet an acceptably mature product though.
Well... If you do a search you'll find that there are actually lots of lawsuits related to open source license violation, mostly around GPL license. Maybe educate yourself first
Your assertion thst copyright doesn’t matter and that people should “get used to it” is… pretty incorrect on the former (surely there are still pending copyright suits on earth circa 2022) and ill-conceived on the latter (I’ll just rob your house while you sleep and you should just get used to it)
> It would be because it's illegal and violates the licenses, desires, and intentions of the thousands of workers who wrote the code in its corpus.
Copilot makes source code much more open, if you think about it. It implements code reuse in a different way than classes and libraries. It offers its skills equally to everyone, skills learned from everyone.
As for the cost of the API - it's expensive to run large language models, I think the price is justified. But there are free models if you like to run your own.
They should in addition release all code generated as a combo of AGPL, GPL, MIT, etc. and put a comment on every usage. Users would then need to license their code accordingly.
For a commercial version, run it on Microsoft's internal code, the code they actually own!
I consider GPL as meaning, "don't make the same app as me using this very source code". Cribbing one method out of quarter million line code-base, hardly seems to be redistributing the source code. Literally, it is, but is there no concept scale? Can we go to an extreme and force anyone with a `catch(Exception e)` line in their Java code to go and prove they did not take it from a GPL (or similarly licensed project)? I think this indicates there is a line, at some point it is enough code where you are recreating the functionality of the software - to me that is the thing that matters. I don't give a crap if you use my GPL code to learn from and use any parts of it to build whatever software, but I do care if you recreate the same software using my GPL code.
I would accept a claim of license violation if someone used copilot to autocomplete so many methods from one specific project that you have recreated that original project.
I still think it is a matter of scope. It can still be the case that a relatively small module is not cool to lift, but I think in this case we are still talking about such small subsets of functionality that it is completely divorced from the original software. Like, I could see it if a specific method were really key in some way to a unique application, a very novel solution to a difficult problem - but if that were the case, how can an AI possibly use that for a training model? In other words, the auto-suggestions of an AI are going to be common coding solutions to common coding problems that the AI has seen hundreds of thousands of times. That individual proprietary GPL, unique and novel solution is not really the stuff of an AI suggestion. In other words, the code that co-pilot is going to suggest is going to be non-unique, generic, and not really specific to the overall application at all.
It’s not illegal, it’s at worst a fancy code search tool that Github has the right to show you the results via the license you grant them when you upload and make public code on Github which is way stronger than other search engines like Sourcegraph have to show public code.
It doesn’t mean you have the right to use any of the code it generates but Copilot itself isn’t illegal in any meaningful sense.
This is definitely not true. When your license requires you bundle said license with any reproductions of the code, and Copilot spits out said code sans license, they are breaking the law.
> We need the legal right to do things like host Your Content, publish it, and share it. You grant us and our legal successors the right to store, archive, parse, and display Your Content, and make incidental copies, as necessary to provide the Service, including improving the Service over time. This license includes the right to do things like copy it to our database and make backups; show it to you and other users; parse it into a search index or otherwise analyze it on our servers; share it with other users; and perform it, in case Your Content is something like music or video.
I don’t think it’s accidental that this product is specifically Github Copilot.
But even then I think this is legal overkill. If you use the search box on Github they will display snippets of code from public repositories without the license. Same as what Sourcegraph does same as Copilot does. Nobody here is arguing ripgrep is violating the license by displaying matches without the corresponding license.
The violations are when that code is incorporated into your own codebase, which is happening in none of those examples. If you copy GPL'd code from GH search with a non-compatible license you are still in violation.
Yes but that’s your problem as a user of the tool. It doesn’t make Copilot itself illegal which is what the person I originally replied to was saying.
Yes if you use a tool to violate copyright it’s copyright infringement. If you prod Midjourney into outputting near exact Starry Night that’s on you too.
So far no one has made a compelling case that Copilot itself is violating copyright.
Code search may show snippets, but it's clearly not separated from the rest of its code base, including the license. It may be your problem as a user of that tool if you pirate snippets out of the search results without honoring its license, but GitHub at least didn't distribute it without its license. A human judge would surely determine that a search result page showing a snippet and linking back to the project wouldn't constitute distributing the code without attribution. Copilot is a different matter. There is no way to know where the code came from, whether it's novel or verbatim copy of someone's copyrighted work. Microsoft _is_ distributing code snippets here sans its license.
Codesearch at least provides users the ability to hunt down any licensing concerns. Unless Copilot start spitting out citations (this snippet was generated based on repositories x, y, z, here are links) the users of the tool have no way to verify if they are in any way in violation of the licenses.
The people protesting aren't a "vocal few"; We're the people who made copilot possible. We are frustrated that our work is being used to profit a massive corporation without any compensation and in a way that we at best did not intend to allow and at worst is in direct violation of the terms we set.
Perhaps a good compromise is to make it opt-out, if it’s not already. Though even this is just pandering to the developer’s ego. AI writing code is a massive boost in giving users power and thus freedom. Of course, we need to make AI itself FOSS, but I doubt a legal case could be made for that. A more productive path is to clone the model like SD did with Dall-e.
Nah, you're definitely the few. Its not a random sample, but an informal survey of my coworkers found no one who would care and generally positive sentiment.
The people who comment on something are disproportionately those who care a great deal.
My point was not that those critical of CoPilot are in a majority, it is that our perspective is important because our labour is what makes copilot possible.
What proportion of its capability is derived from the labor of people who don't like it? I get your point about feeling like an unwilling contributor while github/MS harvests revenue from people who like it. But there's an implication here of being in a critical majority, which I am not convinced is the case.
And if the code generated by copilot was attached to a license that you had to obey? Suddenly your propriety solution must be released as open source or rewritten, because copilot is effectively laundering open source code?
Life's a lot easier when you can just copy whoever did the hard work without crediting/paying/etc for it.
> It's been a massive productivity improvement to our senior devs
I would hate to work at a place where advanced-but-untrustworthy autocomplete would, at all, impact the productivity of a senior engineer.
Not only does this indicate that your senior engineers' productivity is measured poorly (lines of code), but also that your senior engineers are paid to type, rather than to think.
It doesn't indicate either of those things. It's a great tool when you're working on complex code that just reaches the limit of your working memory/attention, and has often suggested good and clean improvements for me. You may dislike it but there are people it helps.
If the tech becomes open (and all indicators point to it being open in the near future) then it will become impossible to shut down. This has already happened with Stable Diffusion and the related model leaks.
People's expectations have already been set by this technology, and they are only going to want more. Also, AI researchers are still publishing their work out in the open for anyone to reproduce.
If there was a Copilot model out in the wild like with Stable Diffusion then this ceases to be a valid question, regardless of the model's legality. All it takes is a single leak or decision by another entity to release their own code generation model.
I tried copilot and found it an excellent way to inject subtle bugs into my code. It always had a seemingly plausible guess, that was never correct, and coding turned into a guessing game trying to spot the bugs it had injected and hoping I’d found them all.
1,267 comments
[ 2.4 ms ] story [ 445 ms ] threadThe author seems to be implying that since Copilot can reproduce the code of open source repository X in certain scenarios there'd be no reason for programmers to learn/use/engage with repository X. But this is silly. Maybe some open source repositories could be tab completed with a little prompting but people will presumably choose to add a dependency instead of tab completing the code of express or something.
(Or I guess more technically, the original authors copyright still applies, and the rights granted to use the work under the license as an exception to the strict limitation under copyright - do not apply...)
Perhaps the author means that there’s a possibility that the programmer to whom the code was suggested won’t necessarily know its provenance and how to engage the community from whence it came. If so, that’s a stronger argument, but I don’t know that it’s the best one they can make.
Writing software just auto-completing from CoPilot would be like trying to write a whole novel with the auto-predictive text on your phone. You could do it, but the results would be non-nonsensical and full of semantic errors. I don't think there's any real 'provenance' at play in either case.
This guy is a literal who, who is really over estimating the value of his open source contributions over a general development tool that can reduce the cognitive load of working in some hairy code bases.
I don't think it's CoPilot that's 'erasing' his, Racket, community.
Reminds me of people who defend AI art with "you can tell it apart from real art" yeah no, at the rate we're moving you won't be able to tell at all in a year or two.
Copilot is an AI stunt, an exploration, trying something new and exciting with very mixed and not-so-useful results.
This lawsuit, however, is just lawyers doing what they do for fun. I guess the retained Microsoft lawyers love it too. Glad to see lawyers having so much fun and profit. But we would all be better off with out so much lawyering, can't they do something more worthwhile?
The license and attribution are stripped from regurgitated copied code snippets from code projects. If the people don’t known which project the code was taken from, how can they one day contribute to that codebase? If the code projects on GitHub are not getting the people who use their code at least aware of the project, that project disappears. Copilot is an interloper who doesn’t even tell you which project the code snippet was ripped off from!!
If I solve a problem and I say: Sure, use my code if you want, but be sure to contribute any improvement back to the community - I wouldn’t be happy seeing a tool spewing it out everywhere. And it’s not even free. In this case, microsoft is literally making money on the back of millions of programmers. And without approval.
Generating snippets of code has nothing to do with a fully functional software package/product/service and an organic community around it.
One might argue that a community could be more easily formed thanks to co-pilot because it increases developer productivity and lower the effort to contribute so OS projects actually benefit from co-pilot. If this sounds far fetched, then probably the first case is also similar.
But they deliberately don’t tell you thst… it’s just a code snippet floating in space like they invented it
If the people don’t known which project the code was taken from, how can they one day contribute to that codebase?
If the code projects on GitHub are not getting the people who use their code at least aware of the project, that project disappears.
Copilot is an interloper who doesn’t even tell you which project the code snippet was ripped off from!!
1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
3. All advertising materials mentioning features or use of this software must display the following acknowledgement: This product includes software developed by the organization.
4. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
5. Use of this source code for the research or training of machine learning models is permitted.
This is no different whether you are honkler or Microsoft (other than in how vigorously or not someone may enforce it).
If you really believe that anything posted to the internet ‘belongs to all’ then I don’t know what to tell you other than you live in a fantasy land where Oracle Corporation does not exist. We might all prefer it if things were that way, but they simply aren’t, and that’s just tough.
Small enough pieces of code can't be copyrighted. No one would support an argument that I violated copyright by using the code "else if {" from some GPL library.
So the question becomes what is the minimal unit of copyrightable code? What if you wrote a nice big function exactly (or almost exactly) the same way as someone else did? Whose copyright are you violating?
People are conflating their open source license with the one they give GitHub when making a GitHub account, but they are two entirely separate and parallel licenses. The former is for other people to use your code, the latter is for GitHub to host your code.
If you don't like it, you are free to host your code on your own servers.
And anyway, as noted the other day about AI, it is often funny to see people not care about (or even enjoy) AI in other fields that they don't work in, but when it comes for their own field, they are suddenly very worried. See programmers on HN who argue for Stable Diffusion but against Copilot, and vice versa with artists on Twitter. As I commented then, it's an act of cowardice to think our own profession should be immune from AI while we enjoy the fruits of AI in other fields [0]:
> Yes, many of us will turn into cowards when automation starts to touch our work, but that would not prove this sentiment incorrect - only that we're cowards.
>> Dude. What the hell kind of anti-life philosophy are you subscribing to that calls "being unhappy about people trying to automate an entire field of human behavior" being a "coward". Geez.
>>> Because automation is generally good, but making an exemption for specific cases of automation that personally inconvenience you is rooted is cowardice/selfishness. Similar to NIMBYism.
We should want AI. That we then try to use outdated models like copyright to enforce holding back human progress is a true shame. In my view, so what if GitHub uses people's code for training data, we are all getting a better product because of that.
[0] https://news.ycombinator.com/item?id=33226515#33228948
For instance, what if I self-host an OSS project but someone puts a mirror on GitHub? Or just uses GH as a remote for their fork? Does that random person accepting the ToS now mean GH has carte blanche to do whatever they want with that IP?
https://docs.github.com/en/site-policy/github-terms/github-t...
4. License Grant to Us
We need the legal right to do things like host Your Content, publish it, and share it. You grant us and our legal successors the right to store, archive, parse, and display Your Content, and make incidental copies, as necessary to provide the Service, including improving the Service over time. This license includes the right to do things like copy it to our database and make backups; show it to you and other users; parse it into a search index or otherwise analyze it on our servers; share it with other users; and perform it, in case Your Content is something like music or video.
This license does not grant GitHub the right to sell Your Content. It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service, except that as part of the right to archive Your Content, GitHub may permit our partners to store and archive Your Content in public repositories in connection with the GitHub Arctic Code Vault and GitHub Archive Program.
--
[1] the practicality³ of this is a different, though related, discussion
[2] because the user is fully informed and can take responsibility for the decision to use the suggestion or not
[3] or impossibility – given the code could be added by someone who doesn't include that attribution/licence information for the system to be able to pass on even if it were designed to
I'll stick to self-hosting instead of using services like GH. Keeps things a little more simple in that regard.
Copilot on the other hand basically defaults to infringing behavior. Users would have to go to great lengths to be sure they aren't infringing on others work.
And if you didn't have the rights to grant the licenses to Github? Then you are in violation of the copyright holder's rights, not Github.
The only remotely plausible, yes-I-have-graduated-fifth-grade argument against Github is that they ought to and certainly do know that huge portions of their users are in fact granting them licenses without the necessary authority. That's an interesting argument we should be having, and instead we're having this inane screaming match by people who have no clue what they're talking about while some of us are sitting here going WTF is wrong with you?
I am discussing the license grant GH includes in its terms. And that doesn't appear to give them a blank check to do anything they want with code those users have uploaded. Certainly not sell it piecemeal.
"It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service."
Throughout the license, the Content is treated as an indivisible unit, and it specifically refers to the forking functionality. Notice that forking...forks an entire repository, licenses included, etc. You can't fork a single file, and you can't fork a region of a file. GH provides that kind of forking.
Copilot is fine-grained forking.
No significant software company is going to permit copilot to be used and potentially poison their code base in unknown ways, now that this kind of copying is in the open and is clearly a significant danger.
Somebody like Black Duck is going to make a lot of money for trial attorneys by tracing how code was created and finding the "hits". That will be joined with log data indicating who used copilot, when they used it, and exactly what copilot presented as the "hit". This entire process will be performed recursively on the "hit", together with classic source analysis, to find out where something is really from.
The bigger companies are really, really serious about not copying outside code except under really strict conditions -- these conditions mostly look like "no you may not, unless you have one of these specific situations". It's no-by-default, even when it looks like it could be a yes.
You've ignored the important words. Copilot is part of the Service.
I don't know, sounds pretty similar to training on ML programs, even if they don't explicitly say "machine learning" in the ToS.
This would, at a minimum, preclude charging for Copilot.
This is missing the point though. Microsoft claims their use of source code for Copilot is fair use. If they are correct about that, licenses don't matter, this EULA doesn't matter, etc. Everyone should be focusing on this claim, arguing about any other detail before that is decided is a waste of time.
> parse it into a search index or otherwise analyze it on our servers; share it with other users
That is the most succinct and most accurate definition of Copilot I've ever seen.
From the article:
> “Dude, it’s cool. I took SFC’s advice and moved my code off GitHub.” So did I. Guess what? It doesn’t matter. By claiming that AI training is fair use, Microsoft is constructing a justification for training on public code anywhere on the internet, not just GitHub.
And:
> when it comes for their own field, they are suddenly very worried
From the article:
> First, the objection here is not to AI-assisted coding tools generally, but to Microsoft’s specific choices with Copilot. We can easily imagine a version of Copilot that’s friendlier to open-source developers—for instance, where participation is voluntary, or where coders are paid to contribute to the training corpus. Despite its professed love for open source, Microsoft chose none of these options. Second, if you find Copilot valuable, it’s largely because of the quality of the underlying open-source training data. As Copilot sucks the life from open-source projects, the proximate effect will be to make Copilot ever worse—a spiraling ouroboros of garbage code.
If I was writing this website, I would delete this sentence, because it is actually really idiotic.
1. This is an opinionated statement, except that there's nothing backing this opinion. This is fearmongering, that GitHub Copilot will get worse unless we sue Microsoft.
2. Lawsuits are not about concerns about a product's creator potentially damaging their own product. Lawyers suing Microsoft don't get a "we're protecting Microsoft from Microsoft's own bad decisions!"
The argument, completely unfounded, is that GitHub Copilot will undermine... GitHub Copilot. So, if you like GitHub Copilot, you should also be on board with suing Microsoft, so that we don't damage GitHub Copilot. What??? Good luck proving that line of argument in a court - you'd get laughed out of the room. Courts don't react well to hazy predictions about mayhem, from the suing lawyers, that have no historical facts to base them on.
I do exactly this, and it does all of bupkis to prevent someone from downloading my code from my gitea, and uploading it to GitHub. In fact, several people have.
https://en.wikipedia.org/wiki/Abstraction-Filtration-Compari...
https://en.wikipedia.org/wiki/Idea–expression_distinction
https://h2o.law.harvard.edu/cases/5004
Most of your code is probably not subject to copyright in the first place, regardless of license.
Copyright is meant to protect “useless” things like poetry and music.
> "You are responsible for ensuring the security and quality of your code. We recommend you take the same precautions when using code generated by GitHub Copilot that you would when using any code you didn’t write yourself. These precautions include rigorous testing, intellectual property scanning, and tracking for security vulnerabilities."
I can't help but recall:
"Linux is a cancer that attaches itself in an intellectual property sense to everything it touches."
- Steve Ballmer, while CEO of Microsoft
They have some really good blow in Redmond.
If anybody could win an award for being coked up and sweaty on stage...
https://www.youtube.com/watch?v=Vhh_GeBPOhs
See also this Domo video that turned it into a song. :) https://www.youtube.com/watch?v=f7ZDH45OAt8
On the one hand stating plainly that mixing in copy-left code and similar can be disastrously dangerous because it is a rampant virus. On the other hand not understanding why people think it might be a problem that their tool could encourage mixing in copy-left code.
According to the opinions about what inclusion of open source code into your projects does, as per the ex-CEO of the company. That seems a bit of a far fetched conclusion, but then, Ballmer did say it.
With "normal" code I can generally see (or figure out) who posted/published it and reach out for explicit permission. It's not uncommon for me to do this.
How is one supposed to do that for the generated stuff? Seems like an awefully hands-off attitude. As challenging as it is, they really ought to be qualifying the input samples of training code before ingesting.
There are some vendors in this space too (BlackDuck comes to mind) but they're $$$ so only within the scope of large corporations.
If anybody has any ideas relating to this type of analysis, I'd be excited to chat. I am working on a project[1] in this space for "Software Composition Analysis" which could potentially overlap with snippet detection for code like Co-Pilot. (We basically just have a big pipeline of analysis jobs that run on code and store the results. I need to update the docs!)
0: https://yangdanny97.github.io/blog/2019/05/03/MOSS
1: https://github.com/lunasec-io/lunasec/tree/master/lunatrace
Seems like best practice recommendation that everyone should apply when downloading a torrent.
"I used the copyright to destroy the copyright."
That sort of plot never works in practice.
I think you need to explain that more. The problem (or at least one problem) being explored here is that by using any code from co-pilot, you are responsible for making sure the licensing is correct. You could unknowingly be using and modifying GPL-licensed code in your non-GPL project, which is a violation if you don't publish your modifications.
We're not talking about expanding copyright, just protecting the existing copyright systems from being trampled by microsoft.
Also, the section breaks and headers and boxes lack obvious rhyme or reason. It scans a tiny bit like a classy version of Time Cube. You keep getting hit with different font sizes and font styles and lines and ribbons and colors and you're not quite sure why.
This is on a powerful PC with a state-of-the-art graphics card.
Can this AI regurgitate the vast majority of the creative aspects of an original/novel piece of software with minimal prompting, to the point where the output code looks mostly and directly cloned to a reasonable person trained in the art?
So "Can this AI regurgitate the vast majority of the creative aspects of an original/novel piece of software" is not the test that, for example, the music industry uses when determining if a sample is infringing. The test there is "is a sample, however small, identifiable as part of a copyright work by a reasonable person trained in the art?"
You can't own copyright in a composition of a single middle c note. But lawsuits have been won for copyright infringement of melodies of 2 bars (fewer than about 16 consecutinve notes). Men At Work lost a copyright case for the flute melody in Land Downunder which is the same as a 90 year old tune Kookaburra Sits In The Old Gunmtree https://www.claytonutz.com/knowledge/2010/february/men-at-wo...
Whether that's done by a flute player or an AI, really doesn't make any difference as far as copyright law sees things.
(Whether copyright law is a "good fit" for source code, and whether it makes sense to apply laws meant for books/literature/music/film to software is a different but very good question. I don't have much in the way of other ideas which take original author's efforts and potential rights to benefit from then though...)
As a long-time open source software developer, I have favored the 2-clause BSD and MIT licenses because they are the simplest licenses that provide me some liability protection. I would release code into the public domain if that didn't increase the likelihood of being sued, whether for liability, or for someone else claiming intellectual rights to code I actually wrote.
I would love to see a lawsuit which requires GitHub to provide their full Copilot dataset.
But of course if the data is already sitting in object storage inside your cloud environment and all you have to do is run some MapReduce jobs to get at it...
Hence: unfair, anticompetitive, intellectual-property-right-abusing behavior. Microsoft GitHub (tm) can prevent anyone else from running the kinds of analysis they do by simple "operational security", while running literally any kind of analysis, model training, etc. they want. Don't like it? But their commercial services and products so you can run Microsoft GitHub (tm) on your very own Microsoft Azure (tm) infrastructure, using Microsoft Visual Studio Code (tm) and Microsoft GitHub Codespaces (tm) so work on _your_ code privately.
Best of all, you can still still take advantage of the huge library of "free" code offered by Microsoft GitHub Copilot (tm) to ensure your private, proprietary codebase still has all of the advantages of Open Source Software, brought to you exclusively by the Microsoft GitHub Platform (tm).
ArchiveTeam has a distributed Github archive project[0]. It's unclear what the status is right now. It seems like a worthwhile idea.
[0] https://wiki.archiveteam.org/index.php/GitHub#Archive_Team_p...
I’ve also built many parallel repo downloaders for CI reasons. You can clone repos all day pretty much with little rate limiting. I haven’t pushed parallelism past 64 per host though
That's not unfair, and not the basis for a lawsuit, it's just business.
> Copilot is not only making money off of open-source, they are making money off of open-source in a way others can't.
Of course! That's why MS paid squillions to buy Github.
most monopolies are completely legal. small town with only one gas station or one grocery store? 100% monopoly, 100% legal.
there are plenty of alternatives to github, paid, free, as a service, and self-hosted. GitHub has a large market share, but no monopoly.
This isn't a company with plenty of goodwill in its sails launching a hip new product. They're not Do No Evil era Google or Apple riding high on the iPod. We can't just pretend there isn't a history.
I posit that they have indeed learned from their mistakes.
1) They train Copilot with repos hosted on github.com. 2) Users who upload code to github.com grant GitHub an explicit license[0] granting GitHub the right to show that code to others. 3) GitHub do not specify which technologies or techniques they may use to show this code to others, meaning they may use any technique they like.
Microsoft have learned. And they have covered their collective asses. Users who host code on github.com agreed to these terms.
Users who don't like their code showing up in Copilot should not be hosting their code on GitHub, because they agreed to have their code delivered to others when they signed up.
[0]: https://docs.github.com/en/site-policy/github-terms/github-t...
> This license does not grant GitHub the right to sell Your Content. It also does not grant GitHub the right to otherwise distribute or use Your Content outside of our provision of the Service[...]
I imagine the crux of this case will be what constitutes "the Service", and whether that includes Copilot. Also whether licensing Copilot counts as selling Your Content.
Which is to say: Microsoft was taking specific action which wasn't a natural consequence of their software, but was being actively enforced to keep out competition. Vendors didn't organically discover consumers weren't interested in getting a PC without Windows and IE, they were prevented from even offering the option lest they be completely denied the ability to offer that at all.
they are not threatening to block access to github.com for all Comcast users if Comcast chooses not to block access to gitlab.com, for example. That would be an illegal activity for a monopoly. simply existing as a monopoly is not itself illegal.
Citation needed. I'm interested to know of examples of how they have stifled competitors. AFAIK Gitlab came of age well after Github had already grown super-large. If Github could have killed Gitlab, wouldn't they have done so?
"GitHub has around 56 million users, whereas GitLab has over 31 million users." - https://radixweb.com/blog/github-vs-gitlab
That is also excluding Azure, Bitbucket, AWS and the plethora of other git repository hosting companies and services. AFAIK, gitlab has more payed (AKA, private or premium) customers than does Github (though I lack a citation there, but that was my understanding when researching this a couple years ago on where to host a private companies code. My take-away then was that Gitlab is actually more popular amongst private companies and Github is more popular for open source projects).
What would be considered Monopoly marketshare? I would suggest that Facebook, Twitter and Amazon have monopolies. If you get shut down as an Amazon seller, that can be 90% of your revenue. Given so many people can leave, and have left Github voluntarily, is implicit evidence that there are very decent alternatives (if not, there would be no alternative and you would be forced to stay with Github for lack of alternatives. That is not the case though, it's easy to just go over to Gitlab. From my perspective, I have no idea how Github can kill Gitlab, I'm curious if there is a vector where Github could use it's network effects to diminish Gitlab, as a specific example. So, how could Github do that?)
Does the law really say you can't include a free web browser if someone else created a paid one?
I don't want to live in a world where potential improvements for consumers get companies sued for antitrust.
https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....
Yes. It's called the Sherman Act, and it's the basis of anti-trust enforcement in the US.
https://www.ftc.gov/advice-guidance/competition-guidance/gui...
<<<The Sherman Act outlaws "every contract, combination, or conspiracy in restraint of trade," and any "monopolization, attempted monopolization, or conspiracy or combination to monopolize." >>>
I know lots of people here don't like it, but it is the law and that was the question; "this" in parent clearly meant "use dominance in one market to gain dominance in another" in grandparent, regardless of whether that's actually the central issue here or not.
For instance, Apple already knew about the portable hardware market and extended their reach into the portable music market via iPod. It used the iPod to reach the music marketplace via iTunes. Used its market dominance to create iPhone and the rest is history.
Maybe that's a personal maxim on your part but there's no law that says if you have dominance (what does that even mean?) in one market you cannot enter another. Think about what you're saying. We'd have no multi-product company if that were the case. What does "dominance" or "market" mean anyway?
As you point out, establishing what dominance is in a market is a tricky thing and it is why governments will investigate any potential signs of it. Now, personally I don't think this is really an issue for Github, or microsoft (the parent company) but let's not pretend market dominance and abuse of dominant position are not a real things.
TLDR: If you have a monopoly in a market, you can't use your position to get a heads up in another distinct market by bundling products together. Ex: Microsoft can't use their monopoly position in the OS market to get an advantage in browsers by bundling IE with Windows.
It's my understanding that this doesn't apply if you don't have a monopoly, and it also doesn't apply unless you're actually bundling the sale of multiple things together.
Doesn't seem to be relevant here IMO.
>or instance, Apple already knew about the portable hardware market and extended their reach into the portable music market via iPod. It used the iPod to reach the music marketplace via iTunes. Used its market dominance to create iPhone and the rest is history.
Look at what you're saying here. Apple had domain experience in hardware and launched a new product (good). Then it used the dominance of that product to muscle into an entirely different market (bad). And the combination of hardware and market has led to Apple being able to extract their tax on half the music market, or whatever they have.
This is exactly what we don't want and why anti-trust laws exist.
Just because there is some existing competition which has a few percent marketshare and technically it's not a monopoly doesn't materially change anything besides a pro forma excuse. Which is why Google have been propping up Mozilla, they want the excuse "but technically there's another browser". However for consumers and the market it doesn't matter that technically there's an option that practically nobody uses.
If we define "open source" as "you can't necessarily use this to train an AI", then Copilot itself is illegal because it's using code without permission.
If we define "open source" as "you can use this to train an AI", then Copilot it legal, but GitHub may be illegally misrepresenting itself as a host for "open source", as the policy it hosts code under isn't truly open-source.
If we define "open source" as "you can't necessarily use this to train an AI" but then GitHub's policy explicitly states "by using us as a hosting provider, you give us permission to use your code to train AI" then they are in the clear. But I doubt they have that clause or at least had it when Copilot was first revealed.
>The Google BigQuery Public Datasets program now offers a full snapshot of the content of more than 2.8 million open source GitHub repositories in BigQuery. Thanks to our new collaboration with GitHub, you'll have access to analyze the source code of almost 2 billion files with a simple (or complex) SQL query.
https://cloud.google.com/blog/topics/public-datasets/github-...
I'm pointing out that this limitation is not meaningful because everyone can access all GitHub hosted source code through BigQuery, where they won't be rate limited.
I'm not comparing BigQuery to Copilot.
Their freemium product is useful to many open-source projects and communities, but you do not have any more rights to use Microsoft's GitHub than you have to Microsoft's Windows.
Is it? Data storage would be prohibitive, but I can see ways to download the entirety of Github in a few weeks/months (assuming my size estimate is accurate).
https://www.gharchive.org/ http://ghtorrent.org/
also available on GCP as a dataset, provided by Github itself: https://console.cloud.google.com/marketplace/details/github/...
I pay GitHub / Microsoft to host my code, and that's all I expect them to do with it, host it, as securely as possible. It sounds like Microsoft are doing more than this so what's your actual deal...if you have one?
GitHub considers the contents of private repositories to be confidential to you. GitHub will protect the contents of private repositories from unauthorized use, access, or disclosure in the same manner that we would use to protect our own confidential information of a similar nature and in no event with less than a reasonable degree of care.
"You have bested me in this epic battle of the minds."
Yeah, Microsoft has bamboozled users for years... that's the whole point.
Speak for yourself. I pay multiple streaming services, music and video, because I prefer creators be able to eat.
We are people, we have ambitions and families. We're not just a faceless corporation.
I want there to be more good music and movies. I want to support artists who create entertainment I enjoy. I go out of my way to buy physical copies of music from artists, wherever possible from the merch table at their shows or from their own websites. I pay to go see movies on the big screen (partly because I like the big screen cinema experience, but also because I understand "opening week revenue" is a key performance indicator for the success of a movie).
I thing copyright is old, outdated, and probably not really fit for purpose for forms of creative work invented in the last 50 years. But I also thing creative workers need to get paid for their effort (juist the same as software developers), and absent a FAANG-style set for companies employing teams of songwriters, musicians, authors, and the like - on FAANG-style salaries, copyright seems to be the option that is working (however badly).
I'll join your "abolish all copyright" crusade as soon as there's an alternative that at least likely to possibly work as well (or better) than the system copyright allows. Just abolishing copyright and erasing the publishing/music/movie/art industries without a transition plan isn't a thing I can support. (At least a transition the artists/editors/producers/writers/etc. I'll admit there's a large chunk of management and legal in the fairly abusive parts of the music industry I wouldn't shed a tear if they all became homeless and destitute overnight...)
I wonder when we'll hear about the first big hack that gets traced back to production code pushed live after CoPilot "suggested" eval(base64decode({webshell}))
(Hell, I've _been_ that incompetent management and engineering leadership at various times over the last few decades...)
I want my code to be used the way people treated text in the old days. There's texts that have been re-written, added to and edited by thousands of people over the centuries and yet they don't come with thousands of pages of attribution notices because why would they?
Whether Copilot is breaching MIT depends on what constitutes a substantial portion, which I am not qualified to rule on.
Also back then, Microsoft had 90%+ share of the PC operating system market and was bundling IE in its operating system. I’m glad the DOJ forced MS to change its ways.
By the time MS bought Nokia, it was already a has been in mobile and the acquisition was a total failure and the game market is competitive.
IMO, if the lawsuit goes to a point where it's likely to be won by copyright owners, Open AI could do the following:
- Use less sensitive code from big corps they have partnership with for training. I bet MS and other have plenty of such code.
- Buy training rights from copyright owners of OSS projects. Many of them have SLAs which allow the owner do much more than the license allows.
- Buy rights to train code, and collect generated code with Copilot from a large number of smaller software companies, likely with exclusions for some sensitive parts. MS has a lot of leverage here (discounts, partnerships, etc).
And if you dig further, the whole repo is mixed with BSD and LGPL "licensed" packages together. It's probably best that CoPilot does not suggest from code that does not have an explicit license stated.
I think originally Tim Davis was complaining about the non public sources for suggestions which Github CoPilot ignored.
This is the same case as with copyrighted photos in newspapers, a paper prints a photo somebody allowed them to use, but then it turns out that person did not have the right to use it in the first place. Did not stop newspapers from printing photos at all.
Here are the search terms: https://github.com/search?q=cs_transpose&type=Code
Whether or not his style his widely known, his code is VERY widely used. Just look up SuiteSparse and try to find all of the downstream uses of it. It is one of the most---if not the most---ubiquitously used set of sparse linear algebra libraries. If you do anything with numerical linear algebra, there's a good chance you at least know what SuiteSparse is, and possibly also know who Tim Davis is.
The bigger issue here is the effect this has on research. Tim Davis not only programmed this library, he did the basic research leading to many of the algorithms in SuiteSparse. He went ahead and released SuiteSparse open source, probably thinking that it would be a good deal for him, provided that its use was properly attributed. Provide a public service in exchange for attribution. This is a reasonable way to get support as an academic. Clearly he has had a large number of industrial collaborations which likely have provided him with a significant amount of funding over the years.
Speaking for myself, if Microsoft has no compunction against behaving this way, I can no longer see the point in publicly releasing research code that I develop using an open source model. Microsoft is clearly telegraphing that they don't give a f** about licensing, although whether that holds if they are litigated against remains to be seen. I think there's an excellent chance many other researchers feel the same way. If you think openness and reproducibility in science is important, this is a problem.
>"Tim Davis gave numerous examples of large chunks of his code being copied verbatim by Copilot, including when he prompted Copilot with the comment / sparse matrix transpose in the style of Tim Davis /."
Copilot regurgitates code and blatantly violates licenses, not even sure what there is to argue about. Not only does it seem straight up illegal and sideline open source communities, I think the next logical step of this is that people who want to avoid having their work vacuumed up and their rights violated simply to move to proprietary software, which would be a huge disaster for open source.
I think there is a good, fundamental legal/societal question of how copyright should apply to AI output. I just don't think our existing copyright structures handle this question well.
Note there is currently a very important case before the SCOTUS that is related to this issue, [1] where the original photographer of a Prince photo is suing Andy Warhol's estate for copyright infringement. The fundamental question is whether the Warhol series of painting are "transformative" enough of the original photo. While there are always gray lines on what "transformative" means, if there is any chance that Warhol's painting are legal and not infringing, I don't see how Copilot could be in the wrong. Copilot's output, even if it contains a substantial amount of the original source, appears to me much more "transformative" than the Warhol paintings are compared to the original photo.
1. https://www.npr.org/2022/10/12/1127508725/prince-andy-warhol...
Is it really better to only draft laws that are clear without courts?
Is that provable?
Badly written law and poor legislators are a problem in any system.
https://www.hup.harvard.edu/catalog.php?isbn=9780674012417
Any higher court ruling might well draw some lines between different domains, but be clear that a ruling against GitHub would almost certainly be a ruling for copyright maximization and against fair use in other respects. So be careful what you wish for.
Copilot is a set of trained weight values in a matrix. There is no source code stored in that matrix. The fact that someone can prompt Copilot with specifically chosen text to generate a short sequence of code that matches a corresponding segment of code used to train the model does not mean that it is somehow "just retrieving" that snippet. It is _generating_ that code, guided by the weight matrix, via pattern-matching based on the chosen textual prompt and surrounding context.
That distinction is significant because one of the primary defenses against copyright infringement in US law is if the derived work is transformative. Copilot is a work derived in part from Github code, but it has unique capabilities far beyond returning short snippets of input code, and the work itself is clearly an extensive transformation of the input data.
This is without even considering whether concrete _outputs_ of the model that happen to match code in a repository used to train it are themselves protected via copyright or not, which is another issue entirely (and not as cut and dried as many folks on here seem to think).
To be fair, record companies were not in the least bit sympathetic. Open source contributors are easier to identify with, though imo it doesn't actually make their concerns more valid
I cannot parse what you are suggesting.
Copilot exists publicly, which also means some copilot-lite thing trained on a smaller subset of repos probably exists privately in many different places. It may not be as good today, but these private instances will improve over time. Since the demand for a copilot-like service exists, eventually a large VC-funded public instance will show up.
In that lens, it is more sensible on the individual level to prepare for a world where copilot thrives than to put all of your eggs in the "ban copilot" basket.
It's been a massive productivity improvement to our senior devs, and I got so used to it that it's an annoyance when Copilot doesn't respond.
It would be because it's illegal and violates the licenses, desires, and intentions of the thousands of workers who wrote the code in its corpus.
You act like Microsoft is trying to do a public service and people are angry about it. The reality is that they're taking billions of hours of work and using it to build a product that only they control.
If they re-released Copilot as FOSS, a lot of the valid criticisms would evaporate.
I wonder how many people on HN would be on the side of the creators if we were talking about content created by Walt Disney and whether pirating was ethical?
Disney used its power to distort copyright laws in a self-serving manner. As an individual you don't have equal power to oppose them (you were supposed to have in a democracy, but lobbying is legal and corporations are people).
Disney is a huge corporation that won't even notice if you pirate a movie, which you may not even have been able to pay for anyway, because of their region-locked twisted maze of distribution and DRM.
OTOH you may be screwed if you're a creator making a living from your work, and a big corp can just take it without paying, launder it through "… in the style of $YOURNAME" query, and say they own it now, because unlike your copyright, their Terms Of Service apply.
Even people who think copyright shouldn't exist may rely on using copyright — against itself. You can't unilaterally say "I don't believe in copyright", because the law doesn't care, but if you license something as copyleft, then the law does care about your anti-copyright license.
People probably would have less of a problem if microsoft breached the license on 100 year old code.
But life of the creator + 70 years?
Reminds me of this:
https://arstechnica.com/uncategorized/2007/07/research-optim...
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1436186
Without copyright, it would be perfectly legitimate and legal for someone to not follow a license (because it would bear no legal weight because of the lack of copyright).
I would argue that open source is best served by strong copyright protections that allow the people created the software to make sure that further changes to it are released back to the community. Weakening copyright law means that it is that much easier for big companies to co-opt some software and not need to release the changes back.
As long as Microsoft can and will wield those laws against me? Darn tootin'
This is also why I fully expect Steamboat Willie to fall out of copyright protection in January 2024 - right on schedule. There's a few countries that have supra-EU copyright terms, but none of them are dealmakers. Nobody is demanding we match Mexico's life+100 terms, for example.
I am 100% on the side of content creators. Regardless of who they are .
The courts tend to take a dim view of theft. Which is what this is.
The article clearly lays out that multiple requests for sound legal basis have gone unanswered . It simply doesn’t exist and Microsoft is operating on a forgiveness vs permission model.
Licensing is 100% about permissions. Clear and explicit enumeration of the permissions (or lack thereof ) for a work.
This class action lawsuit should surprise nobody. It’s a class that is sick and tired of being exploited.
Do not take my work that I contributed with explicit permissions and use it in a way I didn’t grant permission for. Full stop. It isn’t complicated.
You wouldn’t download a car and all that jazz….
https://docs.github.com/en/site-policy/github-terms/github-t...
so much animosity over rights you gave GitHub when you put your code there. "Theft" gimme a break. you license your code to GitHub so they can show it to others. This is separate to the stated license in your code. Nowhere in that terms of service document are the means that the code is shown to users specified.
Also MS themselves don't even claim that training is covered by their terms of service, they claim it is fair use.
Or you are mistaken and "the service" of github, includes all features available on the website including copilot.
Even if you're right and a court rules against them, what's to stop them changing the terms to become compliant?
Moreover terms have been largely unchanged for years AFAIK. If someone agreed to the license years ago, they can't have agreed to copilot use. Also copilot is not a service on their website, it is a separate service and they charge for it, also contradicting the terms.
What does separate service mean? What would Copilot look like if it was not separate?
Elaborate on "not a service on their website", as it is available and listed as a feature "Github Copilot" on their website.
Is the contradiction related to payment for the service, or just because you think it is separate?
Since I thought you were arguing that Githubs own terms prevent them from using the public repositories in Copilot, this is what I argued against.
If you think fair use is involved, then that's the end of the line. If MS claims fair use, then until a court says otherwise, it is. Anyone who thinks their copyright is being violated can get an injunction tomorrow.
Maybe some type of hint shown inside your code when it’s shown at github.com. There is already a text editor.
they claim the training of their model is covered by fair use, but they did not say that was the justification they were using. They don't need to claim fair use.
It's pretty clear from the Terms of Use that they can use code hosted on github.com to provide any service they like, so long as it is a GitHub service. They don't need fair use, they already have the rights to do what they are doing.
Fair use is a major part of copyright law. I do not have to ask permission to use your work.
For you to win in court you have to overcome fair use, you have to overcome innocent infringer, you have to overcome no damages.
Anyone leaving comments saying that there's an obvious way a court would rule on a copyright case involving those 3 things is wrong.
If you use my licensed work then yes, you do need to follow the terms of the license .
The issue of license / contract / copyright is messy. It doesn’t ever seem (in the USA anyway ) to be definitively answered / “solved.”
I chose AGPL v3 only on purpose.
Co pilot and users thereof (so now two levels removed ) utilizing code in whatever work are stealing my work (unless it’s AGPL v3 licensed ). The adding of intermediaries (and the most likely unknown and with no way to know) infringing is going to be very difficult to mitigate. It’s like truly unknowingly buying stolen property,
If I use it under fair use, there is nothing you can do about it.
The fourth factor of fair use is the effect on the value of your work. If I'm not affecting the works value because there is no market for it, because it has no commercial value, you are going to have a very hard time defeating this argument in court.
That changes nothing at all. A FOSS license is, in a legal sense, no different then a proprietary one. If it’s illegal it’s illegal regardless of whether it’s FOSS.
If Copilot was FOSS there'd probably be a few absolutists complaining, but they'd be mostly ignored.
Is "imagining" the same with "copying"? Does copy-right cover learning-right? Can learning and practicing be restricted by the authors? Can visual styles, algorithms and facts be copyrighted? I say no to all of them.
The only implementation that would be allowed is if they had to describe the code via a specification and Codepilot was able to generate it from scratch. It does not do that - it just reorganizes stuff it's already seen, which is a copyleft violation in humans, therefore it should be for a machine created by humans.
The code has had its license violated.
That's for patents not for copyrights. All you need to do is make it a little different, generate until satisfied for patent safe code. Another model can also do the patent checking.
They are making rips of other peoples stuff, selling the contents of peoples “books/movies/songs” sans author attribution or album credits etc… to put it in terms you may be familiar with Vinegar and salt on open wounds. Bad
I'm starting to wonder about these arguments, and whether we've gone into bad faith and hyperbole territory here. Are algorithms subject to copyright? Is it the case that if a GPL work uses a well known algorithm, that GPL work cannot be used as a reference? (Given that algorithms have very limited forms they can take, using an algorithm as a reference is really just copying it. Even translating pseudo-code to code, it's still the same thing).
Can you explain to me how something like using Eulers formula to solve a math problem would not be copyright infringement? A GPL project might use that formula somewhere, but then using that would be a copyright violation?
How about HTML source code, does putting a 'copyright' notice on the webpage make it invalid to then use any of the javascript, even if it has nothing special to do with the domain of the website?
"selling the contents of peoples “books/movies/songs”
Going to this analogy, I don't know if it is really the contents, but more like the first sentence, or even the first few words of that sentence rather than any recognizable subset of that work.
Like, if I have an app that does a spreadsheet, I don't care if you take an implementation of quick sort from my code as reference, but I do care if you use the same and main features of the spreadsheet app that I made.
Anyone saying "in violation" without doing a fair use test doesn't know how copyright works.
Why haven't they uploaded Windows source to Copilot?
Just how much code reproduced violates copyright?
If, instead of Copilot, Bob was giving me code to copy, and it was a AGPL codebase, am I still subject to the AGPL?
The problem is that their product sometimes produces verbatim copies of licensed works, without attaching licensing information. This not only goes against the licenses under which the original authors made these works available. It can also put the product's users in danger of anything from bad publicity to a copyright lawsuit.
CoPilot is a very interesting research project. It's not yet an acceptably mature product though.
Copilot makes source code much more open, if you think about it. It implements code reuse in a different way than classes and libraries. It offers its skills equally to everyone, skills learned from everyone.
As for the cost of the API - it's expensive to run large language models, I think the price is justified. But there are free models if you like to run your own.
For a commercial version, run it on Microsoft's internal code, the code they actually own!
Yeah, the vocal few.
Do you think I give a rats ass that Copilot is duplicating my OS code?
I have to imagine most people are completely ambivalent. Of course I have no proof, I just can’t imagine anything else.
The lines probably fall somewhere along the MIT vs GPL camps.
"Ambivalent" means "of two minds," but I'm going to assume you meant that you're indifferent.
If people are/were indifferent, their licenses should reflect that. They overwhelmingly don't.
Regardless, Microsoft is legally bound to obey the licenses.
I would accept a claim of license violation if someone used copilot to autocomplete so many methods from one specific project that you have recreated that original project.
I still think it is a matter of scope. It can still be the case that a relatively small module is not cool to lift, but I think in this case we are still talking about such small subsets of functionality that it is completely divorced from the original software. Like, I could see it if a specific method were really key in some way to a unique application, a very novel solution to a difficult problem - but if that were the case, how can an AI possibly use that for a training model? In other words, the auto-suggestions of an AI are going to be common coding solutions to common coding problems that the AI has seen hundreds of thousands of times. That individual proprietary GPL, unique and novel solution is not really the stuff of an AI suggestion. In other words, the code that co-pilot is going to suggest is going to be non-unique, generic, and not really specific to the overall application at all.
> If people are/were indifferent, their licenses should reflect that. They overwhelmingly don't.
Apparently, overwhelmingly they do. At least if the licenses used are any indication.
https://github.blog/2015-03-09-open-source-license-usage-on-...
I'm almost entirely certain you're wrong about the desires bit. 99% of the developers who wrote that code won't mind.
This is completely unsubstantiated. I for one would mind microsoft profiting from the closed source code I wrote.
If I’m writing code for a query optimizer, the SQL Server solution isn’t going to magically show up.
It doesn’t mean you have the right to use any of the code it generates but Copilot itself isn’t illegal in any meaningful sense.
I don’t think it’s accidental that this product is specifically Github Copilot.
But even then I think this is legal overkill. If you use the search box on Github they will display snippets of code from public repositories without the license. Same as what Sourcegraph does same as Copilot does. Nobody here is arguing ripgrep is violating the license by displaying matches without the corresponding license.
Yes if you use a tool to violate copyright it’s copyright infringement. If you prod Midjourney into outputting near exact Starry Night that’s on you too.
So far no one has made a compelling case that Copilot itself is violating copyright.
Saves a lot of "hey how do I do this simple thing again?" memory loss issues.
The people who comment on something are disproportionately those who care a great deal.
What proportion of its capability is derived from the labor of people who don't like it? I get your point about feeling like an unwilling contributor while github/MS harvests revenue from people who like it. But there's an implication here of being in a critical majority, which I am not convinced is the case.
Reread your post. Doesn't it sound scary? You are blocked from even thinking and crafting because a specific web service is down.
Even if Google is down you can go direct to Stackoverflow and MDN, and have a choice of information sources.
Also what is "productivity" ... as in features built / month or lines of code / month?
Life's a lot easier when you can just copy whoever did the hard work without crediting/paying/etc for it.
I would hate to work at a place where advanced-but-untrustworthy autocomplete would, at all, impact the productivity of a senior engineer.
Not only does this indicate that your senior engineers' productivity is measured poorly (lines of code), but also that your senior engineers are paid to type, rather than to think.
People's expectations have already been set by this technology, and they are only going to want more. Also, AI researchers are still publishing their work out in the open for anyone to reproduce.
If there was a Copilot model out in the wild like with Stable Diffusion then this ceases to be a valid question, regardless of the model's legality. All it takes is a single leak or decision by another entity to release their own code generation model.
Incorrect suggestions all the time