4 ms·

IANAL, but (2) seems pretty standard in terms of DMCA compliance. As a service provider hoping to remain in the "safe harbor", you generally fulfill all takedo

IANAL, but (2) seems pretty standard in terms of DMCA compliance. As a service provider hoping to remain in the "safe harbor", you generally fulfill all takedown requests and you leave contesting the takedowns on the grounds of error or fair use or whatever to the guy whose video got taken down. Applying editorial discretion or undue delay to takedown requests involves taking on some level of legal risk. So the smart thing to do is automate them. It screws over your legit uploaders if they get DMCA'd, but most companies find that better than picking expensive legal fights.

CWuestefeld 15y ago
I can understand why hotfile wouldn't be able to double-check. But you'd think they could at least put a CAPTCHA on the takedown request, so that an RIAA bot can't rip off a million complaints.
- sp332 15y ago
  Nope, any request that fulfills DMCA requirements has to be obeyed. Any exceptions risks losing safe harbor, and that means you're responsible for all content on your servers.
  - bediger 15y ago
    You're responsible for all content, if you blink at one DMCA takedown? I dunno, that seems a bit extreme. Is this for real, or is this just a lawyerly, overly cautious opinion? Are there cases where this has happened? Does the DMCA itself have wording to this effect?
    - sp332 15y ago
      Normally, you would have to take responsibility for the data on your servers. Legally, there is an exception if you disclaim responsibility for all user-generated content, then you just take the content down and whoever has a problem has to work it out with the user directly. This exception is known as "safe harbor", and there are very specific rules about who gets to claim it. https://secure.wikimedia.org/wikipedia/en/wiki/Online_Copyright_Infringement_Liability_Limitation_Act#Safe_Harbor_Provision_for_Online_Storage_-_.C2.A7_512.28c.29 https://secure.wikimedia.org/wikipedia/en/wiki/Online_Copyri... Basically, as soon as you take responsibility for any of the user-generated content, you can be held liable for all of it. Either you're responsible for the content on your servers or you're not, you can't really pick and choose.
delinka 15y ago
I'd say you are correct. The problem I see in this situation is that Hotfile alleges WB has an incentive to remove content: the removed content seems to get replaced with a link of WB's choosing in an effort to sell media. Keep the automatic "take down" (which should be flipping a bit in your database that hides the content) and flag the replacement content for human review. Without a thumbs up from the human, WB gets no referral link. P.S. The human reviewer really can't put things back up even if they notice it's Linux kernel source code. That puts Hotfile back in the legal hot seat.

CWuestefeld 15y ago

I can understand why hotfile wouldn't be able to double-check. But you'd think they could at least put a CAPTCHA on the takedown request, so that an RIAA bot can't rip off a million complaints.

sp332 15y ago
Nope, any request that fulfills DMCA requirements has to be obeyed. Any exceptions risks losing safe harbor, and that means you're responsible for all content on your servers.
- bediger 15y ago
  You're responsible for all content, if you blink at one DMCA takedown? I dunno, that seems a bit extreme. Is this for real, or is this just a lawyerly, overly cautious opinion? Are there cases where this has happened? Does the DMCA itself have wording to this effect?
  - sp332 15y ago
    Normally, you would have to take responsibility for the data on your servers. Legally, there is an exception if you disclaim responsibility for all user-generated content, then you just take the content down and whoever has a problem has to work it out with the user directly. This exception is known as "safe harbor", and there are very specific rules about who gets to claim it. https://secure.wikimedia.org/wikipedia/en/wiki/Online_Copyright_Infringement_Liability_Limitation_Act#Safe_Harbor_Provision_for_Online_Storage_-_.C2.A7_512.28c.29 https://secure.wikimedia.org/wikipedia/en/wiki/Online_Copyri... Basically, as soon as you take responsibility for any of the user-generated content, you can be held liable for all of it. Either you're responsible for the content on your servers or you're not, you can't really pick and choose.

delinka 15y ago

I'd say you are correct. The problem I see in this situation is that Hotfile alleges WB has an incentive to remove content: the removed content seems to get replaced with a link of WB's choosing in an effort to sell media. Keep the automatic "take down" (which should be flipping a bit in your database that hides the content) and flag the replacement content for human review. Without a thumbs up from the human, WB gets no referral link. P.S. The human reviewer really can't put things back up even if they notice it's Linux kernel source code. That puts Hotfile back in the legal hot seat.