7 ms·

Internet Archive Seeks to Defend Against Wrongful Copyright Takedowns

255 points by jcr 11y ago

acqq 11y ago
The article about that on the Internet Archive written by Brewster Kahle himself: https://blog.archive.org/2016/03/22/save-our-safe-harbor-submission-to-copyright-office-on-the-dmca-safe-harbor-for-user-contributions/ https://blog.archive.org/2016/03/22/save-our-safe-harbor-sub... "We filed comments this week, explaining that the DMCA is generally working as Congress intended it to. These provisions allow platforms like the Internet Archive to provide services such as hosting and making available user-generated content without the risk of getting embroiled in lawsuit after lawsuit. We also offered some thoughts on ways the DMCA could work better for nonprofits and libraries, for example, by deterring copyright holders from using the notice and takedown process to silence legitimate commentary or criticism."
- jcr 11y ago
  And the Internet Archive DMCA 512 Comments are here: https://archive.org/details/InternetArchiveDMCA512Comments https://archive.org/details/InternetArchiveDMCA512Comments And if you want the plain text version of the above: https://archive.org/stream/InternetArchiveDMCA512Comments/Internet%20Archive%20DMCA%20512%20Comments_djvu.txt https://archive.org/stream/InternetArchiveDMCA512Comments/In...
- themartorana 11y ago
  Is the DMCA working the way Congress intended? All we ever talk about here is how it's constantly abused, and used as a club by large organizations over smaller parties. Or, is that pretty much "what Congress intended?"
  - ikeboy 11y ago
    There are two parts. 1. Safe Harbor for websites, means you don't have personal liability as long as you comply 2. Ability for copyright holders to get stuff taken down quickly 1 is working fine, 2 is subject to abuse. The post says that 1 works.
    - bobcostas55 11y ago
      If 1 was working fine, google wouldn't have had to implement Content ID. Which in turn led to even more of 2.
      - ikeboy 11y ago
        My understanding is that Google did that to cut down on the amount of time they spent responding to DMCA requests, not because they were worried about legal issues if they hadn't.
        tobltobs 11y ago
        No, they were forced to implement it, as the copyright holder did not want to discuss with google every time. They wanted the right to shoot first, then ask.
        ikeboy 11y ago
        The original post introducing it explicitly says it wasn't for legal reasons. https://googleblog.blogspot.com/2007/06/state-of-our-video-id-tools.html https://googleblog.blogspot.com/2007/06/state-of-our-video-i... Do you have a source that says they were forced to do it by outsiders? Or any plausible way in which the DMCA required that? Edit: also, content ID is automated checking of uploads. The ability for copyright owners to take down content directly is under a different program, I think.
        pdkl95 11y ago
        That's correct. Also, note that Google's take down system is NOT a DMCA take down request. It does not satisfy the requirements in several places, and it is actually stricter than the DMCA. Google is basically volunteering to allow people to issue take down requests without consequences; if it was an actual DMCA take down, they would have to And yes, Content ID is another system entirely where Google pattern-matches uploads (and old stuff when they feel like it); when hits are found, Google tends to reasign the monetization to the "content owner" that registered with Content ID. Getting that money back has generally been impossible, even in cases where Content ID clearly screwed up (false positive). If you're interested in the history of how Google has been using Content ID, there is an older discussion[1] from a few years ago when Google really started to abuse Content ID hard. [1] https://www.youtube.com/watch?v=bt1ubSVMwaw https://www.youtube.com/watch?v=bt1ubSVMwaw
  - stuaxo 11y ago
    Huh - this would be a great website to check laws, to compare what laws are used for and what we were told they are for, maybe have some sort of score + highlight the best and worst (also trends, too see if scope is being widened or narrowed for certain laws).
walterbell 11y ago
> We are deeply concerned that automated filtering could lead to taking down many materials that are being used in reasonable, legitimate and legally protected ways—especially when the underlying purpose of the complaint is not copyright related but rather an attempt to silence critical speech. How would "notice and staydown" (vs. takedown) work with systems like IPFS which use content hashes? Would there be centrally maintained blacklists against which which all hosting companies would need to screen inbound content?
- acomjean 11y ago
  I don't think it would. since material is supposed to enter the public domain after a time(its been a while as copyrights get longer, but thats another issue), if it was on an automated blacklist, it likely never would removed from the list, making copyright effectively forever.
- pjc50 11y ago
  There already is such a (voluntary) system for child porn.
- gue5t 11y ago
  Even if we assume no user ever has legitimate access to a copyrighted work (which is untrue; take, for a trivial example the author attempting to show off their own work while travelling by downloading a copy of it from their home computer via IPFS), it would be necessary to do this differently based on the copyright laws of the user's country, along the same lines as the Linux "wireless-regdb" which says which wireless frequencies may be used in each country. In general, the reason copyright enforcement is pushed for is not accurate enforcement of laws, but rather maintenance of business models while paving over "minor details" such as legitimate contracts and licenses between copyright holders and others. The digital copyright regime has effectively paved over centuries of intricate law to create a binary of "free" and "nonfree", with no internal distinctions or intermediates. That this is not widely recognized is a sign of how effectively media conglomerates control perception of the issue. You simply can't justify unilateral global takedowns on copyright grounds. Many legitimate countries disagree about copyright rules, and they just aren't so simple as a global "copyrighted?" flag. If you want to be a global moral police, you can justify blocking hashes for those reasons as long as there are no collisions and everyone agrees, but that doesn't sound terribly likely either.
cookiecaper 11y ago
The Internet Archive archives many thousands of user profile pages and other types of user-generated content that the OP may eventually realize is not really wise to have online. Things like the individual message board posting histories of private figures. The Internet Archive has no right to display these things, as they don't have a license from the content's author or the outlet that the content author posted on. There is no public interest served by continuing to display these things, as the poster is a non-noteworthy private person and was in all likelihood just posting nonsense. The OP can contact the message board and probably get them to delete the profiles if they don't have a deletion option already baked into their platform, but what about archives like the Internet Archive? Will they comply with takedown requests for such individual profiles, or fight them as well, pretending that there is some public interest served by keeping it online? Privacy is hard in the internet age.
- arca_vorago 11y ago
  We are talking about abuse of DMCA related to takedowns and copyright, and you are muddling the issue by stretching into privacy. Is privacy covered by DMCA? Is the DMCA something that an OP could use to request takedown of data from a site? If it is, then that would be what you should be speaking about, not grand visions of "how dare they copy and then host things that were on the public internet because some of those things might be privacy-sensitive!" To play devils advocate against myself, I have made mistakes in the earlier days of the internet, that I am glad the archives failed to keep. I do understand that there is a need for privacy friendly user sites, but I am sceptical about what tools are allowed to actually perform this structure. Right now, the internet is a threat to the power that be, which is why we will see an ever increasing attempt to legislate it into the ground. If we allow government corruption to seep into the internet anymore than it already is, the real concern will be one of censorship and propaganda, and user privacy is less to do with publicly posting things you shouldnt, but more to do with the corporate/government merger and data sharing that is going on around us. Loopholes everywhere for suppressing dissidents.
  - icebraining 11y ago
    Is the DMCA something that an OP could use to request takedown of data from a site? Why not? Forum posts are copyrighted at the moment of creation just like any other work, and while one certainly gives a license (implied or not) to the forum, there's no reason why that license would extend to the IA.
brador 11y ago
Solution: monthly full core releases and daily update packs of the complete archive in torrent form. Offload the copyright risk to those more willing to take it and you get to keep doing your friendly neighborhood scraping.
- AnthonyMouse 11y ago
  You might be underestimating the size of their archive. https://archive.org/web/petabox.php https://archive.org/web/petabox.php
  - brador 11y ago
    They could start with a text only archive. Compresses beautifully and would make a great first step.
mirimir 11y ago
It's a nice thought. But this talk of "legitimate", "legally protected" and "wrongful" strikes me as just so much wishful thinking. To be reliably available, stuff must be posted in ways that can't be taken down. It's rather the same distinction as with encryption, isn't it?
cmurf 11y ago
It's fair use for historical and educational purposes. My technical explanation for copyright holders is tough noogies, or look at the hand. I'm just not at all sympathizing with their claims. The Internet Archive is a public good.
drallison 11y ago
The Internet Archive is one of the finest institutions of the Internet. The Archive is a not-for-profit and deserves your financial contributions and support. The proposed change to DCMA Copyright Takedowns rules would help resolve existing problems and ambiguities.
temp1235 11y ago
It seems obvious one should pay damages for making false copyright claims. But.... why should we tolerate copyrights? What is in it for me? I kinda like the idea of people doing paid work. If someone wants a something-for-nothing kind of formula they should pay for it themselves rather than creating an impossible burden for others (if not the whole world) One can get enough funding before creating a work or before releasing it. After release the audience can make donations and/or chose to fund future works. This should be good enough for what we need. Before we chose/prefer global mass persecution until the end of time over the crowd funding formula we should first have a good reason for it. Failure to preserve history for that extra bit of entertainment exploitation is not worth it. Not just because it fails to be entertaining. I can't think of any but there might be a few works important enough for copyright to be an ideal formula but we can't expect it to work on the scale we are having it on right now. At the very least a lack of license should default to something like creative commons. (If I rub a bit of snot on some paper I don't want to own the rights to it.) If enough people want a copyrighted work they are going to get it anyway. The "dream" of artificial scarcity with infinite exploitation has ended. We have to write realistic laws now. Something that doesn't violate basic logic. We the audience would gladly pay for a new season of Star trek. I suppose the fear here is that the audience would have influence on the programming?

255 points by jcr 11y ago

acqq 11y ago

The article about that on the Internet Archive written by Brewster Kahle himself: https://blog.archive.org/2016/03/22/save-our-safe-harbor-submission-to-copyright-office-on-the-dmca-safe-harbor-for-user-contributions/ https://blog.archive.org/2016/03/22/save-our-safe-harbor-sub... "We filed comments this week, explaining that the DMCA is generally working as Congress intended it to. These provisions allow platforms like the Internet Archive to provide services such as hosting and making available user-generated content without the risk of getting embroiled in lawsuit after lawsuit. We also offered some thoughts on ways the DMCA could work better for nonprofits and libraries, for example, by deterring copyright holders from using the notice and takedown process to silence legitimate commentary or criticism."

jcr 11y ago
And the Internet Archive DMCA 512 Comments are here: https://archive.org/details/InternetArchiveDMCA512Comments https://archive.org/details/InternetArchiveDMCA512Comments And if you want the plain text version of the above: https://archive.org/stream/InternetArchiveDMCA512Comments/Internet%20Archive%20DMCA%20512%20Comments_djvu.txt https://archive.org/stream/InternetArchiveDMCA512Comments/In...
themartorana 11y ago
Is the DMCA working the way Congress intended? All we ever talk about here is how it's constantly abused, and used as a club by large organizations over smaller parties. Or, is that pretty much "what Congress intended?"
- ikeboy 11y ago
  There are two parts. 1. Safe Harbor for websites, means you don't have personal liability as long as you comply 2. Ability for copyright holders to get stuff taken down quickly 1 is working fine, 2 is subject to abuse. The post says that 1 works.
  - bobcostas55 11y ago
    If 1 was working fine, google wouldn't have had to implement Content ID. Which in turn led to even more of 2.
    - ikeboy 11y ago
      My understanding is that Google did that to cut down on the amount of time they spent responding to DMCA requests, not because they were worried about legal issues if they hadn't.
      - tobltobs 11y ago
        No, they were forced to implement it, as the copyright holder did not want to discuss with google every time. They wanted the right to shoot first, then ask.
        ikeboy 11y ago
        The original post introducing it explicitly says it wasn't for legal reasons. https://googleblog.blogspot.com/2007/06/state-of-our-video-id-tools.html https://googleblog.blogspot.com/2007/06/state-of-our-video-i... Do you have a source that says they were forced to do it by outsiders? Or any plausible way in which the DMCA required that? Edit: also, content ID is automated checking of uploads. The ability for copyright owners to take down content directly is under a different program, I think.
        pdkl95 11y ago
        That's correct. Also, note that Google's take down system is NOT a DMCA take down request. It does not satisfy the requirements in several places, and it is actually stricter than the DMCA. Google is basically volunteering to allow people to issue take down requests without consequences; if it was an actual DMCA take down, they would have to And yes, Content ID is another system entirely where Google pattern-matches uploads (and old stuff when they feel like it); when hits are found, Google tends to reasign the monetization to the "content owner" that registered with Content ID. Getting that money back has generally been impossible, even in cases where Content ID clearly screwed up (false positive). If you're interested in the history of how Google has been using Content ID, there is an older discussion[1] from a few years ago when Google really started to abuse Content ID hard. [1] https://www.youtube.com/watch?v=bt1ubSVMwaw https://www.youtube.com/watch?v=bt1ubSVMwaw
- stuaxo 11y ago
  Huh - this would be a great website to check laws, to compare what laws are used for and what we were told they are for, maybe have some sort of score + highlight the best and worst (also trends, too see if scope is being widened or narrowed for certain laws).

walterbell 11y ago

> We are deeply concerned that automated filtering could lead to taking down many materials that are being used in reasonable, legitimate and legally protected ways—especially when the underlying purpose of the complaint is not copyright related but rather an attempt to silence critical speech. How would "notice and staydown" (vs. takedown) work with systems like IPFS which use content hashes? Would there be centrally maintained blacklists against which which all hosting companies would need to screen inbound content?

acomjean 11y ago
I don't think it would. since material is supposed to enter the public domain after a time(its been a while as copyrights get longer, but thats another issue), if it was on an automated blacklist, it likely never would removed from the list, making copyright effectively forever.
pjc50 11y ago
There already is such a (voluntary) system for child porn.
gue5t 11y ago
Even if we assume no user ever has legitimate access to a copyrighted work (which is untrue; take, for a trivial example the author attempting to show off their own work while travelling by downloading a copy of it from their home computer via IPFS), it would be necessary to do this differently based on the copyright laws of the user's country, along the same lines as the Linux "wireless-regdb" which says which wireless frequencies may be used in each country. In general, the reason copyright enforcement is pushed for is not accurate enforcement of laws, but rather maintenance of business models while paving over "minor details" such as legitimate contracts and licenses between copyright holders and others. The digital copyright regime has effectively paved over centuries of intricate law to create a binary of "free" and "nonfree", with no internal distinctions or intermediates. That this is not widely recognized is a sign of how effectively media conglomerates control perception of the issue. You simply can't justify unilateral global takedowns on copyright grounds. Many legitimate countries disagree about copyright rules, and they just aren't so simple as a global "copyrighted?" flag. If you want to be a global moral police, you can justify blocking hashes for those reasons as long as there are no collisions and everyone agrees, but that doesn't sound terribly likely either.

cookiecaper 11y ago

The Internet Archive archives many thousands of user profile pages and other types of user-generated content that the OP may eventually realize is not really wise to have online. Things like the individual message board posting histories of private figures. The Internet Archive has no right to display these things, as they don't have a license from the content's author or the outlet that the content author posted on. There is no public interest served by continuing to display these things, as the poster is a non-noteworthy private person and was in all likelihood just posting nonsense. The OP can contact the message board and probably get them to delete the profiles if they don't have a deletion option already baked into their platform, but what about archives like the Internet Archive? Will they comply with takedown requests for such individual profiles, or fight them as well, pretending that there is some public interest served by keeping it online? Privacy is hard in the internet age.

arca_vorago 11y ago
We are talking about abuse of DMCA related to takedowns and copyright, and you are muddling the issue by stretching into privacy. Is privacy covered by DMCA? Is the DMCA something that an OP could use to request takedown of data from a site? If it is, then that would be what you should be speaking about, not grand visions of "how dare they copy and then host things that were on the public internet because some of those things might be privacy-sensitive!" To play devils advocate against myself, I have made mistakes in the earlier days of the internet, that I am glad the archives failed to keep. I do understand that there is a need for privacy friendly user sites, but I am sceptical about what tools are allowed to actually perform this structure. Right now, the internet is a threat to the power that be, which is why we will see an ever increasing attempt to legislate it into the ground. If we allow government corruption to seep into the internet anymore than it already is, the real concern will be one of censorship and propaganda, and user privacy is less to do with publicly posting things you shouldnt, but more to do with the corporate/government merger and data sharing that is going on around us. Loopholes everywhere for suppressing dissidents.
- icebraining 11y ago
  Is the DMCA something that an OP could use to request takedown of data from a site? Why not? Forum posts are copyrighted at the moment of creation just like any other work, and while one certainly gives a license (implied or not) to the forum, there's no reason why that license would extend to the IA.

brador 11y ago

Solution: monthly full core releases and daily update packs of the complete archive in torrent form. Offload the copyright risk to those more willing to take it and you get to keep doing your friendly neighborhood scraping.

AnthonyMouse 11y ago
You might be underestimating the size of their archive. https://archive.org/web/petabox.php https://archive.org/web/petabox.php
- brador 11y ago
  They could start with a text only archive. Compresses beautifully and would make a great first step.

mirimir 11y ago

It's a nice thought. But this talk of "legitimate", "legally protected" and "wrongful" strikes me as just so much wishful thinking. To be reliably available, stuff must be posted in ways that can't be taken down. It's rather the same distinction as with encryption, isn't it?

cmurf 11y ago

It's fair use for historical and educational purposes. My technical explanation for copyright holders is tough noogies, or look at the hand. I'm just not at all sympathizing with their claims. The Internet Archive is a public good.

drallison 11y ago

The Internet Archive is one of the finest institutions of the Internet. The Archive is a not-for-profit and deserves your financial contributions and support. The proposed change to DCMA Copyright Takedowns rules would help resolve existing problems and ambiguities.

temp1235 11y ago

It seems obvious one should pay damages for making false copyright claims. But.... why should we tolerate copyrights? What is in it for me? I kinda like the idea of people doing paid work. If someone wants a something-for-nothing kind of formula they should pay for it themselves rather than creating an impossible burden for others (if not the whole world) One can get enough funding before creating a work or before releasing it. After release the audience can make donations and/or chose to fund future works. This should be good enough for what we need. Before we chose/prefer global mass persecution until the end of time over the crowd funding formula we should first have a good reason for it. Failure to preserve history for that extra bit of entertainment exploitation is not worth it. Not just because it fails to be entertaining. I can't think of any but there might be a few works important enough for copyright to be an ideal formula but we can't expect it to work on the scale we are having it on right now. At the very least a lack of license should default to something like creative commons. (If I rub a bit of snot on some paper I don't want to own the rights to it.) If enough people want a copyrighted work they are going to get it anyway. The "dream" of artificial scarcity with infinite exploitation has ended. We have to write realistic laws now. Something that doesn't violate basic logic. We the audience would gladly pay for a new season of Star trek. I suppose the fear here is that the audience would have influence on the programming?