Meta Tells Court Section 230 Shouldn’t Empower Users And Third Party Apps

Meta has a long history of making sure no one but Meta, or Meta-approved apps, can operate on top of its services. And now, in a new court filing, it’s going to somewhat ridiculous lengths to try to stop one developer from making sure he won’t get sued for creating a tool that automates clicking some buttons in the Facebook UI.

In early May, I had a long post about a kinda wacky lawsuit filed by Ethan Zuckerman against Meta, in which he argued that (among other things) a mostly-ignored clause in Section 230 gave him (and anyone else) immunity for building middleware level products that operated on larger social media sites.

I won’t go over all the background again, but the key part is that Section (c)(2)(B) of 230 says:

No provider or user of an interactive computer service shall be held liable on account of any action taken to enable or make available to information content providers or others the technical means to restrict access to material described in paragraph (1)

With “paragraph (1)” being the lack of liability for any action taken in good faith to restrict access to content deemed to be “obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable.”

While Section 230 is often (misleadingly) discussed as providing social media platforms some sort of unwarranted liability shield, the argument Zuckerman is making is that this clause means that he couldn’t face legal liability if he were to create a tool that would (in this particular example) allow Facebook users to go in and unfollow everything and everyone they follow with the click of a button and allow those users to then share some data with academic researchers.

This lawsuit caused way more chatter among tech policy and law watchers than I’m used to. It was a really novel approach. Some definitely worried that if the lawsuit succeeded, it might make certain bad actors immune from certain lawsuits. But, frankly, I still don’t see that. As I explained, the claims here don’t prevent sites from taking technical measures (blocking, rate limiting, etc.). They only seek to stop Meta from using questionable legal theories, such as CFAA claims of “unauthorized access” to bully developers with threats of huge fines (or even criminal charges).

So, while I wasn’t sure the litigation strategy would work, I am hopeful that it succeeds. It could help establish a more vibrant market for middleware, giving users more control over their own data, as well as the apps and services that they use. In fact, it would bring us back a bit towards the earlier internet world, where building on top of other services was considered normal.

Meta has now filed its motion to dismiss in the case. I’m a little surprised not just by the argument, but by how weak it is. That’s not to say it won’t succeed, as it gives plenty of opportunities for a judge to just toss out the case (judges, especially in the 9th Circuit, seem to really hate declaratory judgment, especially in non-IP cases).

The crux of the argument is this: Zuckerman hasn’t yet created his “Unfollow Everything 2.0” app, and therefore there is no actual controversy, so the case should be dismissed.

The Court should decline Plaintiff’s request to invoke this Court’s limited jurisdiction to issue an advisory opinion about a non-existent tool. Plaintiff’s claims—which are contingent on facts that cannot be known until after he has created and released Unfollow Everything 2.0 and Meta has had an opportunity to evaluate how the tool actually works—are not ripe for review under either Article III of the Constitution or the Declaratory Judgment Act, 28 U.S.C. § 2201. Even if the claims were ripe, jurisdiction under the Act remains discretionary, and there are a host of prudential reasons why exercising jurisdiction would not make sense here. Adjudicating Plaintiff’s claims would require needless rulings on hypothetical applications of California law, would likely result in duplicative litigation, and would encourage forum shopping. Nor is it clear that Plaintiff has a bona fide plan to launch this tool. He has widely spoken about this case in the media, going so far as to characterize it as an opportunity for courts to “shap[e] policy.”1 The Court should reject Plaintiff’s invitation to issue such an advisory opinion and follow the more prudent course of declining jurisdiction.

But, of course, that ignores the very clear chilling effects issue related to this case. Yes, Zuckerman has not released such an app because he saw the legal threats that the creator of Unfollow Everything 1.0 received, as well as Meta’s past legal actions going after companies like Power Ventures. This isn’t a theoretical issue. Meta has shown that it will absolutely engage in lawfare to go after third-party app builders, and thus there is a real controversy and concern here.

Within this, there are some other odd mini-arguments, including the fact that since Meta can change how the “unfollow” button works, the eventual app might work differently, and thus the case is not yet ripe for court.

But, that seems to be a weird way of admitting that Meta has the technical ability to break the app, which no one denies, and no one is challenging. But if the ability to technically break a middleware product means you can never challenge the threat of legal liability… well… that wouldn’t make any sense at all.

Even so, Meta’s lawyers argue that the case should still be dismissed, even if the judge determines that the chilling effects are enough to allow the case to move forward. But they do so by pulling a sleight of hand move to avoid discussing the actual issues:

First, use of Unfollow Everything 2.0’s anticipated automation of Facebook processes would violate Meta’s Terms of Service (“Terms”), which prohibit accessing or collecting data from Facebook “using automated means.”

So what? The whole point of the lawsuit is that in the event that such an app violates such policies, Meta still remains able to use technical measures to stop or block the app, just not legal ones.

Then there’s this very, very strange and legally wrong argument:

Plaintiff’s invocation of section 230(c)(2)(B) of the Communications Decency Act, 47 U.S.C. § 230 et seq., is without merit, as (1) section 230(c)(2)(B) does not immunize Plaintiff from his contractual obligations under Meta’s Terms as a matter of law; (2) it is highly unlikely from Plaintiff’s allegations that he would qualify as an “interactive computer service” provider; and (3) the allegations do not show that Plaintiff’s tool would “restrict access” (rather than “restrict availability”) of content

This paragraph gets so many things wrong, I’m not sure where to begin. We’ll get back to the argument about contractual claims in a minute, but I want to focus on the idea that Zuckerman would not qualify “as an interactive computer service.”

That doesn’t matter.

The law clearly says (and multiple courts have reinforced) that the liability protection applies to “providers or users.” And he is absolutely a user. The fact that he might not be a provider is meaningless. There’s an entire section devoted to this argument, and I don’t understand why Meta thinks it matters, unless they think that the judge (like some amateur Section 230 commentators) will somehow skip over the “or user” part of Section 230.

As for the claim that 230(c)(2) should not bar liability based on contracts liability, I don’t think that’s right and I also don’t think Meta realizes the harm it might be doing in arguing that. Because people in the past have certainly brought “breach of contract” claims against Meta over content moderation decisions, and Meta has relied on 230 to get them dismissed. Be careful what you litigate for, Meta.

Finally, as for the claim that Zuckerman’s app would only “restrict availability” rather than “restrict access,” this seems like a meaningless distinction. What is the actual difference here? How are those different things? It kinda feels like Meta saying “if we call this something different, maybe we can pretend the law doesn’t apply to us.”

Indeed, so much of what Meta is arguing here seems to be about pretending that Zuckerman is saying they can’t use technical means to block his app. But the lawsuit isn’t arguing that at all. It is just focused on making sure that Meta can’t threaten him with a baseless lawsuit for creating an app that is useful to a small group of users, and that simply automates a feature that Meta already makes available to any user directly.

Meta cites some earlier Section 230 cases, such as Alex Berenson’s case against Twitter or the seminal Barnes v. Yahoo. Meta uses these examples to argue that 230(c)(2) liability protections don’t apply to breach of contract claims. However, this twists what actually happened in those cases, which were related to questions around content moderation and whether or not employees of the companies had made additional promises above and beyond the terms of service to do (or, in the Berenson case, not do) certain things.

That is wholly different from what Zuckerman is arguing here: that 230, and its policy of encouraging third-party tools for dealing with content issues online, should make these apps immune from civil liability from the platforms.

One other note on this: in the original lawsuit, Zuckerman and his lawyers from the Knight First Amendment Institute point out (a few times) that he’s hoping to use his Unfollow Everything 2.0 app for academic research. In the Motion to Dismiss, Meta argues that this is meaningless because Meta has policies and offerings in place for researchers to get access to certain data.

Meta supports rigorous and independent research into the potential impact social media services like Facebook may have on the world, and it makes available tools and processes to help researchers gain access to information and analytical capabilities to support their research through a privacy-protective approach. See, e.g., Research Tools and Data Sets, Meta Transparency Center, https://transparency.meta.com/en-gb/researchtools/; Our Approach, Facebook Open Research & Transparency, https://fort.fb.com/approach. Plaintiff does not indicate whether he has considered using these tools or why his research could not be conducted through them.

Who cares? There’s no legal requirement that you use only approved methods of doing research. Why even make this argument?

Anyway, I’m still not convinced Zuckerman will succeed with this legal theory, but Meta’s motion to dismiss seems really off base in multiple ways.