OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

[Posted March 17, 2022 by corbet]

The Open Source Initiative reports on a ruling in the US Court of Appeals reaffirming the meaning of "open source" in a software license.

The court only confirmed what we already know – that “open source” is a term of art for software that has been licensed under a specific type of license, and whether a license is an OSI-approved license is a critically important factor in user adoption of the software. Had the defendants’ desire to license its software as AGPLv3-only been permissible, its claims of “100% open source” wouldn’t have been false and there would have been no false advertising. But adding the non-free Commons Clause created a different license such that the software could not be characterized as “open source” and doing so in these circumstances was unlawful false advertising.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 17:01 UTC (Thu) by nye (subscriber, #51576) [Link] (4 responses)

The most interesting part of this post might actually be the final paragraph:

> There's another interesting aspect to this case. Even though the AGPL has a clause specifically allowing downstream recipients to remove "further restrictions" like the Commons Clause, the court stopped the defendant from doing just that. We'll cover that in our next post.

Looking forward to that one.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 17:18 UTC (Thu) by JoeBuck (subscriber, #2330) [Link] (3 responses)

There's a huge problem with that clause of the AGPL:

If the Program as you received it, or any part of it, contains a notice stating that it is governed by this License along with a term that is a further restriction, you may remove that term.

Suppose someone who is sloppy about licenses takes an AGPL work, combines it with some non-free but legal to distribute software, and includes the AGPL plus the notice giving the license terms for the non-free software. Can a downstream recipient then "remove that term"? Licenses don't have that kind of power.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 19:22 UTC (Thu) by NYKevin (subscriber, #129325) [Link] (2 responses)

What this means is that the distributor has given the recipient a license which is (at least in part) invalid. Unfortunately, that's the recipient's problem. Furthermore, the distributor is in violation of section 10 of the AGPL (v3), which states:

> You may not impose any further restrictions on the exercise of the rights granted or affirmed under this License. For example, you may not impose a license fee, royalty, or other charge for exercise of rights granted under this License, and you may not initiate litigation (including a cross-claim or counterclaim in a lawsuit) alleging that any patent claim is infringed by making, using, selling, offering for sale, or importing the Program or any portion of it.

So whoever actually created the upstream AGPL software can sue the distributor for an AGPL violation, but the ultimate recipient is just out of luck.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 19:52 UTC (Thu) by zwol (guest, #126152) [Link] (1 responses)

This is 100% armchair lawyering but I _think_ the drafters of the AGPL intended the "You may remove" clause to be a self-help option for violations of the "You may not add" clause.

That is: supposing Alice writes some software (all by herself, so she's the only copyright holder) and distributes it under AGPL. Bob makes no changes to the actual code but slaps a clickwrap no-reverse-engineering license on it and distributes it that way. Carol, who receives the software from Bob, can discard the clickwrap and distribute Alice's code under AGPL as-if she had received it directly from Alice. Carol doesn't have to obey Bob's terms while waiting for the Alice v. Bob lawsuit to resolve.

The scenario in the lawsuit is different: the original authors of the software added terms to AGPL _themselves_. I'm guessing the court saw this as the creation of a new "AGPL with extra restrictions" license, and read "You may remove" to apply only to terms added _on top of that_ by third parties.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 18, 2022 0:27 UTC (Fri) by rgmoore (✭ supporter ✭, #75) [Link]

I'm guessing the court saw this as the creation of a new "AGPL with extra restrictions" license, and read "You may remove" to apply only to terms added _on top of that_ by third parties.

IAANAL, but that's the way I read this case as well. The original author can release the software under any license they choose. If they write their license by adding extra terms to an existing license, further distributors aren't free to remove those terms, even if the license it was based on says something about removing additional terms.

In practical terms, I think this was a bad way to write the license. They thought they were saving time by taking an existing license and modifying it, but instead they wound up wasting time with a lawsuit. If they had paid their lawyers to do a good job of writing a license that said exactly what they wanted, it would have cost a bit more up front but would have saved a lot of trouble in the long run.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 20:17 UTC (Thu) by bjartur (guest, #67801) [Link] (15 responses)

So, according to the trial court, the Neo4j Sweden Software License is an open source license. The November 2018 release is the first closed source release, according to the court. Of course, the court is wrong, IMO.

In May 2018, Plaintiffs released Neo4j EE version 3.4, which they continued to offer under an open source license; however, they replaced the AGPL with a stricter license, which included the terms from the AGPLv3 and additional restrictions provided by the Commons Clause (“Neo4j Sweden Software License”).&emdash;the trial court

Also, the court found the false advertising to be material because the word free in free and open source drop in replacement implied that the free version was cheaper than the unfree version. It did not mention if software freedom was material, because the alleged price difference (alleged by whom?) was sufficiently material.

It is undisputed that Defendants made the statements at issue to convince customers to adopt ONgDB over Neo4j EE. Because Defendants misrepresented ONgDB as a free version of Neo4j EE licensed under the APGL, there is no doubt that this price differential (free versus paid) was likely to influence customers purchasing decisions. See Hinojos v. Kohl’s Corp., 718 F.3d 1098, 1106–1107 (9th Cir. 2013), as amended on denial of reh’g and reh’g en banc (July 8, 2013) (recognizing under the UCL that price is material to purchasing decisions). Thus, the Court finds that Defendants’ statements suggesting that customers could obtain a “free and open source drop in replacement” for Neo4j EE were material.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 20:34 UTC (Thu) by juliank (guest, #45896) [Link]

Court being stupid and lawyers going "this worked out ok for us, let's fuck the big picture"

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 22:17 UTC (Thu) by pavon (guest, #142617) [Link]

Yeah, they screwed up those parts. It's particularly bad since the ruling itself hinges on the exact opposite of that background description - that the Sweden License is *not* free and open source, and thus it was false advertising to claim ONgDB was free and open source (since it was they apparently didn't have the right to remove Open Clause restrictions even though the AGPLv3 explicitly states that they do).

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 22:37 UTC (Thu) by KJ7RRV (subscriber, #153595) [Link] (12 responses)

So they ruled that "free software" refers to price, not freedom?

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 23:17 UTC (Thu) by bjartur (guest, #67801) [Link] (9 responses)

Words can be ambiguous. Note that true statements can be false advertising if they are misleading (e.g. if it’s unclear what the word free means). I guess it depends on the audience whether the meaning was obvious in context. Shouldn’t really matter, since the reason an expensive license is needed is precisely because the software is unfree. But the court saved itself the trouble of proving that software freedom was likely to factor into a purchase decision by going with low price, which is well known to factor into purchase decisions. I don’t know if the defendant actually tried to grant invalid AGPL licenses for free, since it did intend to charge the IRS.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 17, 2022 23:53 UTC (Thu) by KJ7RRV (subscriber, #153595) [Link] (8 responses)

Thank you! So they basically assumed free meant "no price" without ruling that, so a company might still be able to call FOSS distributed for a fee "free software"?

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 18, 2022 9:22 UTC (Fri) by fredrik (subscriber, #232) [Link] (7 responses)

The risk of people misinterpreting the meaning of the word free in "free software" as gratis and not libre isn't new. My take on the ruling is that it finds that the word free was at least in part misinterpreted as gratis in this case.

My conclusion is that it would be wise of a company that intends to charge a price when distributing free software to avoid the ambiguity by explicitly stating their intended meaning of the word free. Not only in the license but also in their marketing material.

But IANAL. It would be interesting to get a professional opinion on this matter, in the light of this ruling.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 19, 2022 3:43 UTC (Sat) by KJ7RRV (subscriber, #153595) [Link] (6 responses)

Okay, thank you. So it would take another case to determine what the default meaning of "free software" is?

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 20, 2022 19:32 UTC (Sun) by NYKevin (subscriber, #129325) [Link] (5 responses)

That's simply not how false advertising works. There is no "default meaning" of anything. It's always a case-by-case totality-of-the-circumstances test. The court is not trying to determine some sort of Platonic ideal of what the ad means, they're trying to figure out if consumers would be confused or misled. But such confusion always depends on an elaborate constellation of factors, and cannot be firmly tied down to the meaning of one word or phrase in isolation.

For example, if you display the word "free" right next to a purchase price, consumers might find that hard to understand, but they're (probably) not going to walk away thinking that the price is $0.

Libre vs gratis in advertising

Posted Mar 21, 2022 1:27 UTC (Mon) by KJ7RRV (subscriber, #153595) [Link] (4 responses)

Okay, that makes sense. So an ad that says "Free Software Downloads - $10/program" is probably okay, because "free" can only mean libre there, but "Free Software Downloads" on its own is not (if you actually have to pay), because it could be interpreted as gratis? (I'm just curious, if I ever sell copies of free software I'll make sure I label it clearly, and I know any answer will not be legal advice)

Also, is "Free Software Downloads" false advertising if the software is gratis but not libre (e.g. any of the proprietary "free" apps in app stores)?

Libre vs gratis in advertising

Posted Mar 21, 2022 5:12 UTC (Mon) by mathstuf (subscriber, #69389) [Link]

One could also argue that the *download* is free (a la "free delivery").

Libre vs gratis in advertising

Posted Mar 21, 2022 8:38 UTC (Mon) by anselm (subscriber, #2796) [Link] (1 responses)

So an ad that says "Free Software Downloads - $10/program" is probably okay, because "free" can only mean libre there, but "Free Software Downloads" on its own is not (if you actually have to pay), because it could be interpreted as gratis?

That's what hyphens are for. “Free-Software Downloads” implies that we're talking about downloads of “free software” (a term of art), not software downloads that are free (of charge?).

Libre vs gratis in advertising

Posted Mar 21, 2022 19:17 UTC (Mon) by KJ7RRV (subscriber, #153595) [Link]

I didn't think of the word "downloads" affecting it like that; if a company labels a program as "Free" does it have to be libre, gratis, or both?

Libre vs gratis in advertising

Posted Mar 21, 2022 16:22 UTC (Mon) by rgmoore (✭ supporter ✭, #75) [Link]

I think the key thing is that "free" is a term likely to lead to confusion, since it has two completely different meanings. If you don't specify which meaning you intend, people might legitimately be confused. But this isn't inevitable. If you make sure the discussion is in context, you can make sure people will understand the meaning you intend. If you talk about "free software" in a discussion of licenses and say it meets FSF and DFS guidelines, people will assume you're talking about freedom. If you talk about "free software" but say there will be a charge for advanced features, people will assume you're talking about price.

In this case, you might be able to clarify with some footnotes. If you say something like "Free¹ and Open Source² Software" with the footnotes pointing to FSF and OSI documents talking about your licensing, the meaning would be clear. This is exactly why so many ads that talk about technical stuff include footnotes; they want to make sure nobody can accuse them of being misleading.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 18, 2022 10:01 UTC (Fri) by jd (guest, #26381) [Link] (1 responses)

Well, apparently the ad stated it was "free and open source". This could seem to imply that the "free" refers to something other than the source openness.

The court was also of the opinion that the software's source wasn't open and whether that impacted how they tried to construe what the "free" could have meant is less obvious. They don't seem to have consciously looked at it that way, but that doesn't mean it didn't create some level of bias.

OSI: Court affirms it's false advertising to claim software is Open Source when it’s not

Posted Mar 20, 2022 20:22 UTC (Sun) by NYKevin (subscriber, #129325) [Link]

There is also the https://xkcd.com/1726/ problem. You can tell people that "free" means "libre" all you want, but people are going to interpret it how they interpret it, not necessarily how you intended it. This is, ultimately, the speaker/writer's problem, not the listener/reader's, especially in the field of advertising.