5 ms·
I do not understand why this is unclear: https://en.wikipedia.org/wiki/Section_230_of_the_Communications_Decency_Act https://en.wikipedia.org/wiki/Section_230_
by cft 10y ago
I do not understand why this is unclear:
https://en.wikipedia.org/wiki/Section_230_of_the_Communications_Decency_Act https://en.wikipedia.org/wiki/Section_230_of_the_Communicati...
https://www.law.cornell.edu/uscode/text/47/230 https://www.law.cornell.edu/uscode/text/47/230
"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."
- cloudjacker 10y agobecause lawyers collect fees from emotional claimants even if it gets dismissed in a summary judgement.
- JoshTriplett 10y agoIt isn't unclear, at all. That said, the article doesn't provide the full details of the case. If, for instance, people had reported posts that made specific threats, and Facebook explicitly acknowledged the reports but declined to act (while removing other posts), there might be a case. But if not, and they're just the communication medium, there's unlikely to be a case here.
- dragonwriter 10y agoWell, for one thing, because that's the "Communications Decency Act of 1996", and the US-law claims here arise under the "Anti-Terrorism Act of 2001", and one of the long-established canons of statutory construction is that, where there is a conflict, the newer enactment supercedes the older one. For another, the suit doesn't ask for them to be "treated as the publisher or speaker" of the information, only a knowing collaborator with such a "publisher or speaker", based on their own Data Usage Policy and the manner in which they selectively promote content to particular users.
- kyledrake 10y agoThat's even more ridiculous than 230. Facebook is obviously not a "knowing collaborator" with a terrorist organization. They have hundreds of millions of users, and some of those users are evil. Policing all of them on an individual basis is impossible. That's why 230 is in place. The spirit of this is to try to use a hastily written anti-terrorism bill to do a run-around on 230 for profit, and it's not going to work. Even if the lawsuit succeeds, congress will quickly fix the loophole in a mostly unanimous vote. The alternative would be a horrible tech recession as essentially all social networks and ISPs shut down due to unmaintainable legal risk.
- dragonwriter 10y ago> Facebook is obviously not a "knowing collaborator" with a terrorist organization. I don't doubt that's what the evidence would show at trial; OTOH, the legal system quite deliberately is designed not to operate based on prejudgements about what evidence will end up showing. > They have hundreds of millions of users, and some of those users are evil. Policing all of them on an individual basis is impossible. The plaintiffs' claim is that, based on Facebooks own statements in its Data Usage Policy and its own descriptions of how it selectively promotes content, that it does, in effect, actively police its users, and that the promotion of terrorist content is the result of that active policing. I'm inclined to think that this position is, in fact, incorrect, but I would be less quick to dismiss the complaint as failing to make a colorable legal claim. > The spirit of this is to try to use a hastily written anti-terrorism bill to do a run-around on 230 for profit, and it's not going to work. Perhaps, perhaps the intent of this is to use the Anti-Terrorism Act as a lever to get Facebook to change its business practices to reduce the utility of the platform as a vehicle for terrorists. > Even if the lawsuit succeeds, congress will quickly fix the loophole in a mostly unanimous vote. The alternative would be a horrible tech recession as essentially all social networks and ISPs shut down due to unmaintainable legal risk. I don't think there is near-unanimity in the US Congress that "social networks and ISPs" (even if both were at risk; the particular style of charge here is applicable to the former but not really the latter) are a more favored group than "Israeli victims of Islamist terror groups".
- dogma1138 10y agoIt's not as simple, for example ISP's or "interactive computer service" providers do have to comply with DMCA, and probably quite a few other legislations, meaning that the protection from liability is not absolute. Facebook also isn't just a bulletin board, or an aggregate since they do construct your feed, hence one could potentially argue that they are not a service provider but are the "information content provider" as defined by the Communications Decency Act. That argument could potentially be even made stronger by the fact that facebook does employ selective moderation, they do remove certain content that goes against their TOS including "extremist" content they have deleted groups/pages of ISIS and other organizations in the past, so one could argue that they might be "unwilling" to act in this case (if that indeed is the case). ANAL but It could quite likely not be a case that would be dismissed in a summary judgement, and considering that quite a bit of internal information about what and how facebook moderates, and more importantly how it chooses which information to show to you would potentially be revealed in the lawsuit there's a good chance that Facebook would likely go into a settlement, probably not a billion dollars but also probably not a small penny either.
- dragonwriter 10y ago> It's not as simple, for example ISP's or "interactive computer service" providers do have to comply with DMCA No, they don't. The DMCA gives them an optional procedure, which, if followed, provides a "safe harbor" exemption from liability they might have under other laws. To the extent that the CDA has already extinguished liability, there is no benefit from complying with the DMCA.