6 ms·

The biggest impact I could see this happening would be sinking auto-DMCA bots, since without a human in the loop there isn't any consideration of fair use at al

by zyxley 11y ago

The biggest impact I could see this happening would be sinking auto-DMCA bots, since without a human in the loop there isn't any consideration of fair use at all.

pdabbadabba 11y ago
This is something I've been wondering about. Do we know the extent to which this process truly is entirely automated, and not so much as glanced over by an actual human? I would think that even a cursory human review of "infringements' flagged by a bot could satisfy the legal standard in this opinion, and would add little overall cost for the takedown mills.
- DanBC 11y ago
  RIAA sent 100,000 takedown requests to a single website. It's unlikely a human did much scrutiny of that. http://arstechnica.com/tech-policy/2015/09/sharebeast-the-largest-us-based-filesharing-service-has-its-domain-seized/ http://arstechnica.com/tech-policy/2015/09/sharebeast-the-la... And there are sites that routinely get clearly bogus takedown requests. Google deleted 200million search results in 2013. https://torrentfreak.com/google-deleted-200-million-pirate-search-results-in-2013-131112/ https://torrentfreak.com/google-deleted-200-million-pirate-s...
comex 11y ago
Notably, the decision mentions automation in passing as a potentially acceptable method, but the example given is a program determining that two videos are almost identical, a very far cry from the standard used by such bots today: > We note, without passing judgment, that the implementation of computer algorithms appears to be a valid and good faith middle ground for processing a plethora of content while still meeting the DMCA’s requirements to somehow consider fair use. Cf. Hotfile, 2013 WL 6336286, at *47 (“The Court . . . is unaware of any decision to date that actually addressed the need for human review, and the statute does not specify how belief of infringement may be formed or what knowledge may be chargeable to the notifying entity.”). For example, consideration of fair use may be sufficient if copyright holders utilize computer programs that automatically identify for takedown notifications content where: “(1) the video track matches the video track of a copyrighted work submitted by a content owner; (2) the audio track matches the audio track of that same copyrighted work; and (3) nearly the entirety . . . is comprised of a single copyrighted work.”
- Natsu 11y ago
  This would seem to be rather weak, so at least they're saying this only as dicta. We've already seen cases where people use public domain content and then end up using such algorithms to claim the content away from other people, which would seem not to be covered by the above, though perhaps in that case they would say that they should have known of their lack of ownership before turning the bots loose. And it would seem to spell trouble for the people who just grab the top X results for certain search keywords and then DMCA them without even looking at the sites. At the very least, I would like to see them put the burden of proof on the person filing the DMCA takedown as to precisely where and how they formed their belief in good faith whenever the takedown is wrongly filed. It's not like anybody else is in a position to testify as to their state of mind and if they're truly responsible parties, they should have no trouble in accounting for what factors somehow mislead them. The courts can also request details on their accuracy: how many such notices have they filed, how many of those have been putback, etc. to decide whether or not they're actually behaving responsibly. Naturally, I would presume that any party unwilling or incapable of providing such details was, by that very fact, irresponsible. Anyone not watching what they're doing is being careless.
  - JupiterMoon 11y ago
    We have to remember that bots are written by humans and in this case merely implement a business process. There is no reason that the owner of the bot can't store their own logs of what take down requests are sent and with which version of their bot and therefore (given a well written bot) this should give them an adequate log of the reasons behind the take down notice. It seems to me like this ruling will cause the required sophistication of bots to be greater rather than precluding their usage.
    - Natsu 11y ago
      Every one of them is also, nominally, signed by a lawyer. There are words for lawyers that don't have records of the things they've signed and 'responsible' isn't among them.

by zyxley 11y ago

The biggest impact I could see this happening would be sinking auto-DMCA bots, since without a human in the loop there isn't any consideration of fair use at all.

pdabbadabba 11y ago

This is something I've been wondering about. Do we know the extent to which this process truly is entirely automated, and not so much as glanced over by an actual human? I would think that even a cursory human review of "infringements' flagged by a bot could satisfy the legal standard in this opinion, and would add little overall cost for the takedown mills.

DanBC 11y ago
RIAA sent 100,000 takedown requests to a single website. It's unlikely a human did much scrutiny of that. http://arstechnica.com/tech-policy/2015/09/sharebeast-the-largest-us-based-filesharing-service-has-its-domain-seized/ http://arstechnica.com/tech-policy/2015/09/sharebeast-the-la... And there are sites that routinely get clearly bogus takedown requests. Google deleted 200million search results in 2013. https://torrentfreak.com/google-deleted-200-million-pirate-search-results-in-2013-131112/ https://torrentfreak.com/google-deleted-200-million-pirate-s...

comex 11y ago

Notably, the decision mentions automation in passing as a potentially acceptable method, but the example given is a program determining that two videos are almost identical, a very far cry from the standard used by such bots today: > We note, without passing judgment, that the implementation of computer algorithms appears to be a valid and good faith middle ground for processing a plethora of content while still meeting the DMCA’s requirements to somehow consider fair use. Cf. Hotfile, 2013 WL 6336286, at *47 (“The Court . . . is unaware of any decision to date that actually addressed the need for human review, and the statute does not specify how belief of infringement may be formed or what knowledge may be chargeable to the notifying entity.”). For example, consideration of fair use may be sufficient if copyright holders utilize computer programs that automatically identify for takedown notifications content where: “(1) the video track matches the video track of a copyrighted work submitted by a content owner; (2) the audio track matches the audio track of that same copyrighted work; and (3) nearly the entirety . . . is comprised of a single copyrighted work.”

Natsu 11y ago
This would seem to be rather weak, so at least they're saying this only as dicta. We've already seen cases where people use public domain content and then end up using such algorithms to claim the content away from other people, which would seem not to be covered by the above, though perhaps in that case they would say that they should have known of their lack of ownership before turning the bots loose. And it would seem to spell trouble for the people who just grab the top X results for certain search keywords and then DMCA them without even looking at the sites. At the very least, I would like to see them put the burden of proof on the person filing the DMCA takedown as to precisely where and how they formed their belief in good faith whenever the takedown is wrongly filed. It's not like anybody else is in a position to testify as to their state of mind and if they're truly responsible parties, they should have no trouble in accounting for what factors somehow mislead them. The courts can also request details on their accuracy: how many such notices have they filed, how many of those have been putback, etc. to decide whether or not they're actually behaving responsibly. Naturally, I would presume that any party unwilling or incapable of providing such details was, by that very fact, irresponsible. Anyone not watching what they're doing is being careless.
- JupiterMoon 11y ago
  We have to remember that bots are written by humans and in this case merely implement a business process. There is no reason that the owner of the bot can't store their own logs of what take down requests are sent and with which version of their bot and therefore (given a well written bot) this should give them an adequate log of the reasons behind the take down notice. It seems to me like this ruling will cause the required sophistication of bots to be greater rather than precluding their usage.
  - Natsu 11y ago
    Every one of them is also, nominally, signed by a lawyer. There are words for lawyers that don't have records of the things they've signed and 'responsible' isn't among them.