NY Times Publishes A Second, Blatantly Incorrect, Trashing Of Section 230, A Day After Its First Incorrect Article (original) (raw)

from the what-is-wrong-with-the-times? dept

Last week we wrote about the NY Times having to issue a whopper of a correction on a giant front page of the Business Section, totally false claim, saying that Section 230 “protected hate speech” online — which they later had to edit to note that it was actually the 1st Amendment that protected such speech (and the article leaves out that it’s actually Section 230 that allows websites to remove hate speech). Coming from a paper that, just weeks earlier, had published an editorial mocking politicians for getting 230 wrong, this was kind of embarrassing.

Even more embarrassing, though, was the day after the NY Times had to totally correct that false article, they ran another blatantly wrong op-ed about Section 230, this one published by Jonathan Taplin, who two years previously had published another op-ed at the NY Times that completely fabricated a bunch of blatant lies about how YouTube and Google operate. You would think that would be enough for the NY Times to maybe think twice about having him publish another op-ed, especially about Section 230 a day after the paper got called out for getting the Constitution wrong. But, nope. Taplin got to publish his anti-Section 230 op ed with no problems, until the NY Times felt the need to issue a correction on that one too.

An earlier version of this article misstated the law containing a provision providing safe haven to social media platforms. It is the Communications Decency Act, not the Digital Millennium Copyright Act.

Taplin, if you don’t recall, is an old-school movie and music producer, who a few years ago started getting renewed attention by blaming Google for basically all the ills in the world. He constantly popped up with facts-optional arguments, that seem to get lots of attention because attacking Google is the in-thing, whether or not it’s based on any facts.

Here, again, Taplin gets basically all the facts wrong. While he purports to be an experts on copyright law (a dubious claim in itself), since everyone’s focused on Section 230, he decided to jump in and pretend to be an expert on that (though, as the correction shows, he confused CDA 230’s protections with the DMCA 512 safe harbors, which are quite different). Even post correction, Taplin’s article is just… hilariously wrong.

After the El Paso massacre, the 8chan website briefly went offline when Cloudflare, the network provider, banned it. But it was only a matter of time before someone else agreed to support the site: By Tuesday morning, 8chan was back online, spewing a toxic mix of hatred, violence and QAnon conspiracies. Message boards like 8chan aren?t the only place where these things are spread: In March, when there were two mass shootings at mosques in Christchurch, New Zealand, the shootings were live-streamed on Facebook and then viewed millions of times on YouTube.

Of course, what Taplin leaves out is the rather Herculean efforts that Facebook and YouTube went through to block that video — while also trying not to block legitimate news reports about the shooting. And you know what made that effort possible in the first place? Section 230 of the Communications Decency Act. But, to Taplin, Section 230 is to blame.

Though it may seem that there is little that platforms and politicians can do to stop the spread of online hatred, a great deal could be accomplished with one simple tweak to the existing Communications Decency Act: revise the safe harbor provisions of the law.

Well, here’s the thing: The Communications Decency Act does not have any safe harbor provisions. This is the fundamental problem in the NY Times allowing Taplin to write his facts-optional nonsense in which he confused the DMCA 512 (which does have safe harbors) with CDA 230 (which does not) — and then pretending that merely replacing DMCA 512 with CDA 230 fixes things. A safe harbor, such as what’s found in the DMCA, says that if you take certain actions, you are protected. CDA 230 requires no such thing — it just says that internet service providers cannot be held liable for third party content or for moderation choices.

And, again, it’s the protections of 230 that made the actions that Facebook and YouTube took in response to the Christchurch shooting video possible in the first place. But, Taplin’s got a beef with these companies, and he’s going to scratch that itch, no matter how nonsensical.

A safe harbor provision of a statute or a regulation specifies that certain conduct will be deemed not to violate a given rule. For social media platforms like 8chan, Facebook and YouTube, the most important of these is Section 230 of the Communications Decency Act, which states, ?No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.?

That’s not a safe harbor.

the second one should be: When Johnson & Johnson removed all the Tylenol from American stores in the wake of a poisoning scandal, it did so because of the liability it would face if anyone got hurt. But 8chan, Reddit, Facebook and YouTube are totally protected from being sued for content on their networks because of this unique set of laws created at the beginning of the internet.

What a weird and dumb analogy. Johnson & Johnson faced liability because its products were poisoned. Products where it controlled the manufacturing and the supply chain. That is not the case for the service providers Taplin names. Besides, as noted in the correction to the other NY Times article, the fucking 1st Amendment protects that content. But, of course, Taplin can’t actually admit that he’s really trying to blame the 1st Amendment.

The safe harbor laws were created for what is known as passive (or neutral) intermediaries.

This is a lie. This is such a lie that the NY Times itself called it out as a lie just a couple weeks ago. The safe harbor was not created for “passive (or neutral) intermediaries.” Literally both of the authors of CDA 230 — Chris Cox and Ron Wyden — have directly said this is not true. Section 230 was created in direct response to the Stratton Oakmont v. Prodigy ruling, in which a judge said that because Prodigy heavily moderated its forums to keep them “family friendly,” it took on liability for any content it left up. Cox and Wyden have both made it clear that the entire point of 230 was to enable sites to moderate content however they saw fit, to encourage the ability to create “family friendly” environments online.

In other words, it was created, explicitly, to enable non-passive intermediaries. Taplin is lying. I’d say he could just be misinformed, but this information has been out there long enough, and Taplin has been so wrong, so often, that it’s difficult to see how he could be merely misinformed. At some point you have to wonder if these constant misrepresentations are intentional.

Verizon, for example, is a passive intermediary platform: It makes no attempt to edit or alter the bits flowing through its fiber optic cables. Facebook and YouTube, however, are active intermediaries; they present you with content different from what they present to me. They filter pornography and jihadist videos off their networks using artificial intelligence. As such, they should not be shielded from liability by safe harbor laws in the same way that Verizon is shielded.

This is also bizarre. Given what we just said above, without a CDA 230, then Facebook and YouTube would be incentivized not to filter pornography or jihadist videos. This entire piece is internally inconsistent. Taplin first says that because of 230 these sites don’t have incentive to moderate (even though they do) and then says that because they moderate, they should lose the protections that make it possible to moderate.

How the hell did the NY Times think it was appropriate to publish this utter nonsense?

Even though Facebook was able to use A.I. to block 90 percent of the Christchurch streams after it identified the video, last year Mark Zuckerberg testified to Congress that it might take five to 10 years to perfect these tools. But society cannot wait five to 10 years ? we need to stop these videos now, and banning toxic content must become the highest priority at 8chan, Reddit, Facebook and YouTube.

Dude. In the previous paragraph you argued that these sites shouldn’t get 230 protections because they moderate. Now you’re saying that they shouldn’t get them unless they automatically stop all bad stuff? From paragraph to paragraph the argument changes, unless you consider “Facebook and YouTube are evil” is the only argument Taplin can make.

Some may argue that deciding what counts as toxic video content is a slippery slope toward censorship. However, for the past 75 years, since the first television broadcasts, the Federal Communications Commission has been able to regulate offensive content on television.

This is not a serious take. This is something someone with literally zero knowledge of how any of this works would make. It does not deserve to be placed on the op-ed pages of the NY Times and is a total embarrassment. First off, the FCC only has a mandate over the airwaves, because they are a (somewhat) limited resource, that the US government handed out for the public benefit — and, as part of that exchange, there was a promise that those who received that spectrum, would not use it for obscenity. Second, the idea that the FCC has ever been able to successfully regulate “offensive” content on TV is laughable and shows a startling lack of understanding of the history of the FCC and its attempts to respond to claims of offensive content on TV. Third, there are strong arguments that the FCC’s determinations on offensive content do, in fact, violate the 1st Amendment, and are very much part of a slippery slope towards censorship.

Finally, it’s quite stunning to see a Hollywood producer coming out and supporting the FCC’s attempts to regulate content on the airwaves. That used to be the kind of thing that Hollywood — back decades ago when it actually was a force for free speech — used to fight against.

I believe we can all agree that mass murder, faked videos and pornography should not be broadcast ? not by cable news providers, and certainly not by Facebook and YouTube.

Then you better get to work repealing the 1st Amendment. Because that’s what protects this content.

Since broadcasters do not have the protection of ?safe harbor,? they engage in a certain level of self-regulation, to avoid being sued.

No, that’s not what’s going on at all. Broadcasters pick and choose what limited content goes online, because it’s not an open platform that anyone can post to. They self-regulate because they are aiming for whatever content they put out to reach a mass audience because they can only produce so much content. It’s not because they’re afraid of the FCC or because they don’t have CDA 230 to protect them.

And there is no reason to believe that the largest corporations in the world ? Google, Apple, Facebook and Amazon ? would behave differently from CBS, Fox, NBC or ABC.

Other than that this is an apples to orangutans comparison. CBS, Fox, NBC and ABC pick and choose exactly what content they broadcast. They have 24 hours a day to fill, and that’s it. The other services are open platforms where anyone can post anything. And that’s an entirely different situation. That Taplin seems to think these are comparable frankly does not speak well of Taplin’s understanding of anything related to the internet.

In the past, Google and Facebook have shown that they can pivot quickly and that they already have the technology to keep certain content off their platforms. There is almost no pornography on Facebook or YouTube because of sophisticated tools that search for and prevent such uploads. And since 2017, both companies have actively removed jihadist videos. But it took the right incentives to get them to do both of those things.

Note that it did not take a change to CDA 230 to make Facebook and YouTube decide to remove jihadist videos. He’s also wrong. This started way before 2017. Also, in taking down “jihadist videos,” Facebook and YouTube have also been deleting and hiding evidence of war crimes. So, hey, if Taplin’s cool with deleting evidence of war crimes, good on him. But some of us can recognize that there are serious tradeoffs to decisions that some people erroneously think are easy calls.

Changing the safe harbor laws so that social media platforms are held accountable for the content their users post would incentivize Facebook and YouTube to take things like the deep-fake video of Nancy Pelosi and the Christchurch shooting videos more seriously.

Except that both of those videos are protected by the 1st Amendment. So, no, it wouldn’t create liability for those platforms. Indeed, creating such liability would create incentives for platforms to bury their heads and not learn of such videos, a la the Stratton Oakmont ruling.

In the wake of the Christchurch shootings, Prime Minister Jacinda Ardern of New Zealand remarked about the social networks: ?They are the publisher, not the postman. There cannot be a case of all profit and no responsibility.?

A quip is not reality. Again, even Taplin admitted in his piece that Facebook blocked the vast majority of those videos. To then pretend that they don’t block these videos because of a profit motive is… nonsensical.

In the end though, Taplin is going to Taplin. The real question is why would the NY Times allow this blatant joke of an op-ed to be published in the first place? Especially after Taplin’s previous op-ed was filled with similar nonsense, which it got called out for, and the Times itself got mocked for misunderstanding 230 just a week after its own editorial pages mocked politicians for misunderstanding 230? The NY Times should admit that publishing Taplin was a mistake. But it won’t, because it’s the NY Times.

Filed Under: cda 230, dmca 512, jonathan taplin, misleading, section 230
Companies: ny times