> That is to conclude, by having a contract (a license is a contract), you lose your rights to pursue any IP/copyright claims (the two are exclusive in French law).
GPL was written by lawyers for the American system of law and France is not part of the USA. The systems of law are different and therefore the implications of GPL-licensing code are different.
I'd be very surprised if France is the only country that treats software licenses like this. The basis of the USA system of law is completely different from even other European courts, let alone non-Western systems of law. Assuming you can write a license for use in the USA and enforce it the world around is quite foolish.
> GPL was written by lawyers for the American system of law and France is not part of the USA.
The important thing is GPL (and other licenses) are contracts. So violations of there terms is a breach of contract and in turn clearly handled by the court responsible for handling breaches of contract.
The interesting part is why it's not (majorly) handled as a breach of contract in the US?
The trick is following or similar sentence (qouted from GPLv3):
> will automatically terminate your rights under this License
Which means of the breach the license contract you lose the license contract and in turn can be sued for copyright violation.
Now the gotcha is not everything you write in a license agreement is necessary valid in exact the same you write it, so now what happens if automatic contract termination is not valid and contracts in such case e.g. exists in-between valid and invalid until conflicts are resolved or similar?
At this point you would still have a contract, one which grants you the right to use given software "just not in given way" and in turn wouldn't it be a case for courts responsible for contract breaches?
Now I know nothing about French Copyright law, but the idea that laws (against auto termination) which would exists to uphold fair competition and prevent abuse might majorly negative affect copyleft licensed open source makes you think.
>Now the gotcha is not everything you write in a license agreement is necessary valid in exact the same you write it, so now what happens if automatic contract termination is not valid and contracts in such case e.g. exists in-between valid and invalid until conflicts are resolved or similar?
You'd have to dig to find the details of the agreements and what they mean for GPL'd software, but it's not as simple as "France is not part of the USA". I'd be surprised if the authors of the GPL didn't take the differences into account. They aren't fools.
Not sure if naïveté or innocence, but I think you are overestimating the GPL efforts by a lot.
RMS may not be a fool but he is still American (and not a lawyer). I don't think he was focused in solving licensing/copyright issues in the rest of the world in the early 90s - and even late efforts for GPLv3 didn't change it that much.
> GPL was written by lawyers for the American system of law and France is not part of the USA.
The GPL also has been ruled a valid contract in the US, and the US also has the rule that with a contract in place, conduct authorized conditionally by a contract but alleged to be done outside of its authorized conditions must generally be pursued as breach of contract.
Would love a citation where this was actually litigated.
Edit: And to be clear here, I'm not doubting you, I am actually quite interested in software license/copyright litigation. Having had long discussions over which is better "BSD" or "GPL" and discussing the litigation of the BSD vs AT&T established some solid case law on both the legitimacy of, and the way the BSD license was viewed. But for a long time there hasn't been good litigation of the various GPL flavors that actually went to court and came back with a decision that was appealed and affirmed. Would love to add such citations to my collection.
This is not the case. In the US the Copyright Act supersedes all other common law and State statutes, this is written into the Act itself. See: Title 17 USC Section 301(a). It is not possible to sue for breach of contract in relation to a copyright claim.
Nope. 301(a) is talking about things like common law copyright. The notes to that section from the Cornell LII [0]:
> Nothing in the bill derogates from the rights of parties to contract with each other and to sue for breaches of contract; however, to the extent that the unfair competition concept known as “interference with contract relations” is merely the equivalent of copyright protection, it would be preempted.
If you think about it, if what you said were true, then you couldn’t make any contracts regarding copyright works and expect them to ever be enforced. Disney wouldn’t work as a concept. Sounds a bit extreme, is a bit extreme.
>Assuming you can write a license for use in the USA and enforce it the world around is quite foolish.
In this case there is no reason to suspect it (the GPL or any software license) is not enforceable in France, you just have to go about it the right way and sue in the contract law courts.
As a specific wrinkle of the French system, if you sue under one system (copyright or contract), if you fail you can't sue under the other. In most countries you would not face such a restriction.
It's amazing to me that this has not come up before. Has a software license never been litigated as a copyright violation in France before? Strange.
This is a travesty. The French law makes no sense.
The victim of a tort is barred from justice simply for not being expert legal scholars or for hiring a bad lawyer.
What's even crazier, The GPL isn't even in French https://www.gnu.org/licenses/translations.en.html
, and French courts don't accept English evidence, so the idea that there was a contract is absurd. (and how can France even allows "contracts" where one party doesn't even notify the other that they agree to it?!)
Also, Orange violated the contractual rights of every user of their software, not just the rights of the author. So why can't other users sue Orange? Or maybe they can? They should!
In a lot of western countries that are not the US, software isn't copyrightable as-is, only as a description of its functionality or something like that (not a lawyer, might have been the patent vs. copyright thing).
This difference means that a lot of what a litigation-centric society would assume about software (perhaps not copyright, but patents) doesn't apply elsewhere.
That said, the goals of the legal systems tend to converge: it was about protecting ideas so the people investing in it stand a chance to see a return. I wouldn't be surprised if there simply is a different system with a different name but with the same goal (yet being incompatible with a global blanket software copyright claim). I have only studied the subject for a small group of western european countries (France not included) and most of them not being based on common law makes quite the difference.
I can't think of any Western country in which software is not copyrightable as-is. There are some where software is not patentable, but that is a very different thing that is not relevant to this story.
Most countries are a signatury to the Berne Convention, so in general, copyright is pretty similar in those countries.
Patents are a different story and different countries have different pantent laws.
Because it’s a horrible mess of confusion between copyrights and patents, and even says so, so it’s hard to see what the author was thinking even posting it.
Reverse engineering is allowed under French law, there are specific legal exemptions around that.
It's common to see EULA/license/contracts with clauses that prohibit reverse engineering, these clauses are void in France and many other jurisdictions.
In the EU, and probably elsewhere, there are entire engineering shops of which the one and only purpose is to reverse engineer competitors’ features and figure out a legal way to get around patents and replicate those features.
No interoperability there.
Some people I know do that for a living in the auto industry.
There is some variation within the EU, with different national laws. In French law at least you’re right: reverse engineering is allowed only for interoperability and you’re not allowed to distribute decompiled code.
Rather it seems like they might have to sue you for breach of contract, instead of breach of copyright.
Per the article the court didn't rule that licenses were unenforceable, they ruled that they were enforceable via contract law and that the plaintiffs were incorrectly attempting to enforce them via copyright law.
But as the article also points out, the different ways of enforcement typically result in drastically different levels of compensation (read: incentive) with contract law being judged on missed profits, which is often zero for FOSS libraries.
During the piratebay case in Sweden, products that has never been for sale and would never be for sale were discussed in terms of missed profits. The argument that the Swedish courts accepted was a hypothetical sale of said product using the closest similar product, with an added multiplier for the fact that the product would never be allowed to be sold in the first place.
It is hard to know if the french courts would accept a similar argument, but as the author I would definitive argue it. Depending on what the FOSS library do and what a similar product would cost if acquired through legal purchase, the number could easy reach into millions of "lost profits".
But can you enforce click-through licenses in France?
If Entr'Ouvert had chosen the alternative route and sued as a contract violation, then Orange would have claimed that they never accepted the contract and the matter could only be decided by the copyright court. But (according to the article) you only get one shot at the litigation. So I guess you can never enforce the GPL in France. Maybe?
The point of the GPL is that, if you don't accept the license, then you don't have any rights to distribute the code and doing so would be a copyright violation. It's not a EULA that has extra restrictions on what you can do.
I don't think taking and distributing a library is comparable to a click-through license. There's a relatively simple case to be made that you could have only copied the code if you had accepted the contract.
Not that Orange wouldn't have tried to claim what you are saying, but as a layman, the case for contract law sounds stronger than copyright infringement.
I'm totally unfamiliar with French Civil law; can contract law ever be punishable with any prison time? In the US copyright law is (at least in theory) a criminal offense, punishable with prison time, but my understanding is that breech of contract is not.
The implications, if true, would be that if you license something in France, and start selling copies in flagrant violation of the license, you cannot be threatened with criminal charges?
> I'm totally unfamiliar with French Civil law; can contract law ever be punishable with any prison time?
Private parties in most countries can’t prosecute criminal charges anyhow, and limitations on the mechanisms of pursuing civil [0] law claims by private parties don't generally have any effect on criminal prosecution by the State, since even with the same common name for an area of law, the precise definitions in criminal law are different than civil law.
[0] used here in contradistinction to “criminal” as an area of law, not to “Common” as a system of law, since both distinctions have been relevant in the thread.
> That doesn't seem right. Does that mean when I buy any proprietary software and then reverse engineer and leak the source code, they can't sue me?
Yes it's legal in France if it's for "interoperability purposes" (so depends on your specific case). Basically any DRM is breakable legally due to this exception.
Reverse engineering is always legal as well and no contract signed can remove that right.
My understanding is that no contract can prevent you from doing what you want with stuff you buy, including disassemble, study, repair, modify, or any use in a non-intended way. IP still apply so it doesn't give you the right to share any information you may find, except for "interoperability purposes".
> he basis of the USA system of law is completely different from even other European courts, let alone non-Western systems of law.
This suggests that it's meaningful to divide legal systems into “Western” vs. “Non-western”.
The most meaningful divide is rather “civil law” vs. “common law”. Japanese civil law is probably closer to any continental European civil law than any of them are to say the U.K.'s or U.S.A.'s common law.
This is interesting new information to me, but I looked it up and Scotland indeed has what many consider a mixed system, quite unique since the U.K. is a unitary state.
I was already aware that Scotland uniquely for the U.K. has the unus testis, nullus testis principle which is ubiquitous in civil law systems, but I had no idea of how far reaching it was.
I think many civilizations independently invented it.
The innovation that law is written down by lawmakers is not that hard to duplicate. If anything, the opposite happened and high frequency of common law in the world is purely because the British Empire was the largest empire to have ever existed and all states that utilize common law descend from the British Empire.
Until the British Empire, common law existed only on the British Isles and the rest of the world had civil law.
Should've gotten some Canadian lawyers involved in GPL. Due to the combined heritage of English and French (Quebec) Canada has both Civil law (French) and Common Law (English).
While this is a true factoid, civil law in Canada is restricted to Quebec, and is ultimately subordinate to the constitution, which is interpreted in a nearly identical manner to American law, in the common law style. The civil law influence is mostly restricted to Quebec in areas of provincial jurisdiction (family law, small claims, etc.) The Copyright Act is federal and would get the common law handling.
(Approximately the same situation is true in the USA with Louisiana, whose state laws also have a form inherited or at least strongly influenced by the Napoleonic code, ultimately, but since harmonized to fit into a common law system.)
Quebec never signed off on the 1982 constitutional reform, which created a mechanism for amending the constitution and finally fully separated Canada from the UK government. Until that point, the constitution of Canada was technically part of UK law. Not entirely unlike the devolved Parliament of Scotland today. The UK Parliament could amend it, in theory unilaterally. (It was long-standing convention not to do so except when requested, by the 20th century.)
So Quebec's signature was not required, legally, for the current constitution to be imposed upon them. (Long ago, Quebec's elected delegates did approve of the original Constitution Act in 1867.) Per the amendment formula, each province including Quebec has veto power over any further amendment that requires unanimous consent (affecting distribution of government powers, etc.) Which is now something of a stand-off. Getting an amendment that triggers that clause passed, seems all but impossible in contemporary Canada politically.
If you’re familiar with the US Constitutional Amendment process, not all states have to sign onto the amendment for it to go into affect. And when it does, even the non-signers are bound by it.
US Constitution had to be ratified by a minimum of 9 states (out of 13 total) before it went into effect. In practice, the Continental Congress made it operable with 11 signatures; the remaining 2 happened after that.
It was a single province, which at the time had around a third of the population indeed.
Their signature was not required for the constitution to enter in effect. It was desired (it still is) but since the refusal was (and still is) mostly political, it is not regarded as a major impediment.
> I couldn't really imagine one of the 13 colonies not signing the declaration or constitution.
Canada never declared independence from the UK. That's probably the difference in thinking. The constitution was not revolutionary. It was a reorganization, and that reorganization was carried out according to UK law (which was equivalent to Canadian law as a UK colony) at the time. The law did not require the consent of the provinces to amend the constitution at that time. In fact, the law allowed the UK parliament to simply rewrite Canada's constitution unilaterally. So they did, upon the request of the Canadian federal government after years of consultations in Canada to propose the amendments the UK rubber-stamped, when they renounced all further claim to that in 1982.
If you want an analogy, some US states rejected the prohibition amendment, but they were bound by it anyway. Established constitutional law allowed it to be imposed on those states.
> If you want an analogy, some US states rejected the prohibition amendment, but they were bound by it anyway. Established constitutional law allowed it to be imposed on those states.
Except it was a constitutional amendment. And the process to create such an amendment is described... in the Constitution itself, which, naturally, every state signed.
Yes. Exactly the same was true in 1982 in Canada. The UK Parliament commissioned a board in 1867 of elected representatives from the regions of what now compromise eastern and central Canada. They drafted a constitution. Quebec was involved and signed off on the Constitution Act 1867. That constitution was passed into law by the UK Parliament.
Per the established constitutional law operating in Canada in the 1980s, following the convention of the previous centuries, the government petitioned the UK government to amend the constitution yet again, this time moving the process of amendment to the Canadian legislatures and fully severing the connection.
Constitutional rule in Canada begins when the British first appear and have jurisdiction, as far back as the 1600s in the eastern parts of the country. It was never interrupted. The Crown has always reigned. 18th century treaties and proclamations in UK law are an inherited part of our unwritten constitutional body. It was simply a different historical course from the USA, but not fundamentally less democratic. IMO.
(Although I risk sounding like a mad Canadian nationalist at this point, I would note that under the modern constitutional system, Quebec could just exit the confederation. That is not a right of American states.)
> That's... weird. I mean, I couldn't really imagine one of the 13 colonies not signing the declaration or constitution.
The US Constitution went into effect on ratification of 9 of 13 states. Virginia and New York were numbers 10 and 11, but ratified it before a government under the new constitution was elected. North Carolina waited about a year later to ratify it, and Rhode Island even several months after that (it ratified it only after the United States threatened to embargo them), and consequently were not part of the first government under the new constitution.
A number of US states did not ratify some constitutional amendments and they still entered in effect. Minimum requirement is two-thirds of the states not unanimity.
I meant "political" in the sense of "it's a gesture designed to cause some political effect or to be used in the future for political negotiations" - maybe not the best word but couldn't think of a better one when writing that.
The word "political" has changed meaning recently to exclusively refer to discussion of minority rights and issues. If you asked (Americans at least) with no context, whether people wanted politics out of government, probably 50% would say yes.
But then I assume that I could sue a French alleged-copyright violator in the USA, and if I win, the infringer is banned from selling that product in the USA?
Yes, that would probably work. USA jurisdiction still applies to software sold in the USA, even if it's originally written in France. Things might get more complicated when a French company hires remote workers, though.
In a global marketplace though, if you can't sell your GPL-derivative in one large market, the fact that you can sell it in a smaller market may not be much comfort.
.png){kind=link}
> That is to conclude, by having a contract (a license is a contract), you lose your rights to pursue any IP/copyright claims (the two are exclusive in French law).
GPL was written by lawyers for the American system of law and France is not part of the USA. The systems of law are different and therefore the implications of GPL-licensing code are different.
I'd be very surprised if France is the only country that treats software licenses like this. The basis of the USA system of law is completely different from even other European courts, let alone non-Western systems of law. Assuming you can write a license for use in the USA and enforce it the world around is quite foolish.