I'm confused why twitter even has to do anything here.

For example: http://chillingeffects.org/dmca512c/notice.cgi?NoticeID=1860...

Is it really illegal to link to infringing material?

Oh.. that's right. http://www.2600.com/news/112801-files/universal.html

Wow, that's so stupid. Why don't they just strip the hyperlink part and leave the URLs un-linked? It'd be a nice legal way to comply while disagreeing.

I do understand copyright applying to avatar and background images. If someone is using someone else's artwork or photography for either one of these, I can see a DMCA notice coming in and being honored.

I suppose there could also be some sorts of material where 160 characters could exceed quoting for fair use. Such as haikus.

I do not understand all the takedowns because of mere mentions of pirate sites, or inclusion of their names, or even linking.

Has linking really become illegal? That was one of the nightmare scenario people were worried about some years ago, but which was dismissed as something that would never happen.

I completely and totally contest that a URL to an actual site is a copyright violation. (Barring the extreme cases such as a URL containing a haiku, but that doesn't seem to be the case in any of the incidents listed.)

Certainly if Twitter had to remove a link to a link to The Pirate Bay (http://chillingeffects.org/dmca512c/notice.cgi?NoticeID=1860...), so would Google. I'm not a lawyer, but it seems that Twitter is being "extra cooperative" here; that's not totally unreasonable for a smaller company with limited resources.

It's not really new. Google has been getting (and posting) DMCA notices asking them to remove links for years now. I think there's even something in the DMCA that deals with search engines, but I don't remember the details offhand.

Can someone explain to me how links to copyright content come under DMCA? Twitter is most definitely not hosting the copyright content? Why are they even at the receiving end of takedown requests?

Chilling Effects has a FAQ under each takedown notice :

>Question: Why does a user-generated content site get DMCA takedown notices for links users have posted?

>Answer: Many copyright claimants are making complaints under the Digital Millennium Copyright Act, Section 512(d), a safe-harbor for providers of "information location tools." These safe harbors give providers immunity from liability for users' possible copyright infringement -- if they "expeditiously" remove material when they get complaints. Whether or not the provider would have been liable for infringement by users' materials it links to, the provider can avoid the possibility of a lawsuit for money damages by following the DMCA's takedown procedure when it gets a complaint. The person whose information was removed can file a counter-notification if he or she believes the complaint was erroneous.

http://chillingeffects.org/dmca512c/notice.cgi?NoticeID=1807...

> Can someone explain to me how links to copyright content come under DMCA?

I'd surmise it is because to non-technical users, the distinction between a "link" to the content and the "content" itself is something they don't understand. From their point of view, the "link" is the content because when they click the link, the content appears.

The vast majority of atty's filing dmca notices would fall into the non-technical user category.

Profile images and backgrounds can be infringing and are hosted on Twitter. Enough people hijacked the identities of celebrities and businesses to need the Verified Accounts program after all.

Google takes down links under DMCA, too.

We've automated the whole DMCA process here at Oregon State University. We have Python scripts that parse the takedown emails, a Python suite that parses network logs to associate it with a user, emails appropriate people, disable network access if they do not respond, complete with django webapp and api.

That's how safe harbor works. As a carrier, you do no judgement or filtering or you lose the status. The user can say "they're wrong", get reinstated and it's then on the copyright holder to bring suit against the user and need a legal order to remove the content again.

Is there an API to Chilling Effects? Would love to do some fun things on this dataset.

Yeah I personally would love to see some basic data breakdowns per site, service, region, etc.