SFC v. Vizio remanded back to California state courts

What's GREAT is that it will send a massive message "if you don't comply, you could get hit from anywhere". And compliance is so damn EASY! (At least, it is if you're half-way concerned about being legal!)

But it'll send shockwaves through the "cheap who cares if it's legal" market. And good!

I don't know about America, but even with this case being American, it'll have a big impact in Europe, and it really will scare off any European company from dodgy far-east suppliers, because like I said non-compliance will be considered wilful, and it WILL HURT.

(And then Americans will benefit because, being faced with being locked out of the European market, European compliance will lead to American compliance because, well, it's just EASIER.)

Cheers,
Wol

It is not the first one, there have been a similar case in France:
<https://lwn.net/Articles/353923/>

If they can't comply with any two licenses individually and together, they should be staying away from publishing any software that uses such license combinations. If the practical sense is that there is less licensing violations, that's a good outcome.

Note that there is a crucial difference: in this case (and two others) the contract argued is between the seller and the buyer, and does not involve the copyright holder.
This argument is that when you buy a software, the seller must make sure it is properly licensed.
In particular, the buyer has standing to sue, which would not be the case for a copyright case.

But that does not mean the GNU GPL is a contract between the copyright holder and the seller as asserted
by L. Rosen (interalia).

(to give an analogy: if you buy a windows laptop and you find out it run a bootleg version of windows instead of a properly licensed one you asked, you can sue the seller for contract breach even if MS will not bother. The ruling was that the above apply even if you replace 'windows license' by 'GNU GPL'.)

Perhaps, but there's ample first-party precedent: The AGPLv3 itself is not strictly compatible with the GPLv3, even though they're both GNU licenses with GPL in the name. You can combine AGPLv3 and GPLv3 files within the same project but the combined work cannot be distributed wholly under the terms of the GPLv3; it remains a hybrid with different parts subject to different terms based on their origin. (See <https://www.gnu.org/licenses/license-list.en.html#AGPLv3.0>, second paragraph).

As for the trademark violation, saying that the license is "AGPLv3 with Commons Clause" is arguably a descriptive use of "AGPLv3". This is not quite the same as advertising the work as being available under AGPLv3 (without mention of any modifications) while adding the incompatible Commons Clause to the actual license. On the other hand, it's hard to argue that it wouldn't be clearer with a completely different name, and the only reason I can see to mention AGPLv3 at all is to try to take advantage of its name recognition.

I agree that standardized terms are generally better, but even if they had used a standard integration clause I suspect it still wouldn't work out the way they intended. We're not talking about a case where the licensor put the work under AGPLv3 (unmodified) and had a separate agreement alongside which imposed additional terms. (The grsecurity case comes to mind as an example of this: the code was GPLv2 but there was a separate agreement with each customer for access to future releases which penalized sharing it. Though that's a borderline case since the penalty for sharing the code was only losing access to future releases, something the GPL did not obligate grsecurity to provide to anyone.) In that situation AIUI a standard integration clause would negate the separate agreement. Here, however, the license was "AGPLv3 with Commons Clause", which is shorthand for a single license which includes both the terms of the AGPLv3 and the terms of the Commons Clause.

Also, the standard integration clauses I've seen only supersede *prior or contemporary* agreements, and perhaps specify how consensus is shown for amending the terms in the future (e.g. in writing with the signature of a specific officer of the company or their successor). I'm not sure a clause prohibiting *future* changes altogether, contrary to the wishes of both parties, would be upheld.

I feel it's important to point out that compliance actions, like reverse engineering, is hard, time consuming and has very few people doing this work.