Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Censorship Google

Google Receives Takedown Request Every 8 Milliseconds 155

Posted by Unknown Lamer from the can-we-just-fix-copyright-already dept.
Via TorrentFreak comes news that Google is now being asked to remove one million links per day (or an average of one takedown notice every 8ms). In 2008, they received one takedown request approximately every six days. From the article: The massive surge in removal requests is not without controversy. It’s been reported that some notices reference pages that contain no copyrighted material, due to mistakes or abuse, but are deleted nonetheless. Google has a pretty good track record of catching these errors, but since manual review of all links is unachievable, some URLs are removed in error. ... The issue has also piqued the interest of U.S. lawmakers. Earlier this year the House Judiciary Subcommittee had a hearing on the DMCA takedown issue, and both copyright holders, Internet service providers, and other parties are examining what they can do to optimize the process. In the meantime, the number of removal requests is expected to rise and rise, with 10 million links per week being the next milestone.
This discussion has been archived. No new comments can be posted.

Google Receives Takedown Request Every 8 Milliseconds More

Google Receives Takedown Request Every 8 Milliseconds

Comments Filter:

  • Google don't be evil (Score:3, Funny)

    by Anonymous Coward on Wednesday August 20, 2014 @11:46AM (#47713157)

    No one forces you to provide a search engine that accepts illegal content. Just screen everything before it goes into the index or don't host it, as simple as that.

    • If the entertainment studios want certain content taken down they can spend the 5 minutes it takes sending the request.

      • Re: (Score:2)

        by ls671 ( 1122017 )

        "5 minutes it takes sending the request"

        More likely, they have automated the whole process and it takes them milliseconds to send the request to several parties.

        • I thought this was a Yoda joke. Yes, I need a cheap laugh today.

        • Including a list of 5000 other websites that have no relation whatsoever to the website which is hosting the infringing content.

        • "Earlier this year the House Judiciary Subcommittee had a hearing on the DMCA takedown issue, and both copyright holders, Internet service providers, and other parties are examining what they can do to optimize the process."

          Note who wasn't present. Anybody representing the public.

      • Except that it's really easy and fast to put up something that is legitimately infringing on a legitimate copyright (assuming you philosophically aren't completely opposed to copyright, and hence consider some copyrights legitimate - I consider copyrights morally legitimate if they're less than 28 years old, for example). Either you drop the idea of copyright enforcement, or you allow rapid-fire DMCA notices, or you make it easy to file life-destroying lawsuits.

        The DMCA is a real problem, but so is the

    • Faulty logic (Score:5, Insightful)

      by s.petry ( 762400 ) on Wednesday August 20, 2014 @11:52AM (#47713221)

      Your statement is based on an absolutely false assumption. You really don't have to look hard to find that most requests have nothing to do with illegal content. The overwhelming majority of the take down requests are for censorship purposes.

      • so take your homepage down. that's how silly it's getting. geez

      • Your statement is based on an absolutely false assumption. You really don't have to look hard to find that most requests have nothing to do with illegal content. The overwhelming majority of the take down requests are for censorship purposes.

        And even if they weren't, a lot of things get censored in the process that never should be. Censorship, even by accident, is Not Okay.

    • No one forces you to provide a search engine that accepts illegal content. Just screen everything before it goes into the index or don't host it, as simple as that.

      Uh, hey how about instead of trying to "screen" all content (by the way, that's a cute word you used there for censoring), perhaps you should realize no one forces you to use Google. It's as simple as that.

      Oh, and one more thing. Go ahead and try and define "illegal" on a global scale. Good luck.

      • Oh, and one more thing. Go ahead and try and define "illegal" on a global scale. Good luck.

        There has never been an easier task.

        Here's a list of what is "illegal" on a global scale:
        1. Everything

        • Oh, and one more thing. Go ahead and try and define "illegal" on a global scale. Good luck.

          There has never been an easier task.

          Here's a list of what is "illegal" on a global scale: 1. Everything

          Yup. But hey, look on the bright side. Indexing will be so fast you won't even be able to measure it.

    • No one forces you to provide a search engine that accepts illegal content. Just screen everything before it goes into the index or don't host it, as simple as that.

      Sure no problem. We'll just lookup each file uploader and check to see if they're authorized to distribute the material in question...what's that you say? It's impossible to verify the identify of most uploaders? Well we can at least check every file against the master database of copyrighted material...wait, you're saying there is no such database? Hmm...maybe your plan needs a little more work.

    • Re: (Score:2)

      by Lennie ( 16154 )

      The problem is context.

      On one site the same content is illegal on an other it is legal.

  • Google should be wary (Score:5, Informative)

    by Opportunist ( 166417 ) on Wednesday August 20, 2014 @11:47AM (#47713161)

    It would not be the first top dog search engine that disappears into obscurity because all it displayed anymore were paid-for links, ads and other crap the powers that are considered "agreeable".

    • It would not be the first top dog search engine that disappears into obscurity because all it displayed anymore were paid-for links, ads and other crap the powers that are considered "agreeable".

      Be wary? Oh that's rich.

      These monopolies have billions in cash reserves to run them profitless for a very long time.

      Like decades.

      In other words, Too Big To Fail is here to stay, regardless of the market captured or lost. Better get comfortable with the elected overlords. They're not going anywhere.

      • These monopolies have billions in cash reserves to run them profitless for a very long time. Like decades.

        Aside from the rather questionable assertion that Google is a monopoly, the company's cash reserves are nowhere near that large, or, rather, the company's expenses are much larger than you believe. Last I heard, Google has cash reserves of ~$60B (which, note, aren't actually cash; you don't leave that much capital sitting idle), and annual operational costs of about $40B. How long Google could continue to operate with hugely decreased revenues depends on just how far the revenues declined, and how much econ

    • You didn't stop using the old ones because the new ones were better?

      Did you stop using Google when the ads were not clearly different from the search results? Cos that was a big deal for a while.

      Maybe you did, but a shitload of people who obviously disagree did not.

      Figure out your fundamental point and come back.

  • or they could just NOT do it (Score:5, Interesting)

    by cellocgw ( 617879 ) <cellocgw&gmail,com> on Wednesday August 20, 2014 @11:51AM (#47713197) Journal

    Instead of having software automatically remove every alleged infracting page, how about having the software automatically send a notice back informing the complainant of a lack of credible evidence, and dropping all the takedown notices into some summer intern's Inbox?

    I mean, jeez...

    • Re: (Score:3)

      by afidel ( 530433 )

      They can't do that, the DMCA very clearly says that the provider must remove the infringing material, then the poster can challenge the takedown, failing to remove the content as requested removes their safe harbor and opens them up to copyright infringement claims with statutory damages of $100,000 per violation, never going to happen.

      • Re: (Score:2)

        by AmiMoJo ( 196126 ) *

        Can someone explain why websites being targeted don't submit bulk count-notices in response? There is no penalty for that either, other than inviting a lawsuit. Since most of the sites are not in the US and have no assets there they presumably don't care about US lawsuits anyway.

    • If they asked for evidence, they'd become liable for any infringement. This is all controlled by the Digital Millennium Copyright Act of 1998. Under DMCA, when they receive a notice, they have to remove the content. If the other side sends them a counter-notice, they have to put it back up. If the complainant then notifies the carrier that they are filing suit in federal court, the carrier takes the content own again. If they choose not to follow this procedure, if they set themselves up to judge the evi
      • In short, the DMCA is a so incredibly retarded and unbalanced law (it give full power to the author of the request with no practical penalty for making a false request) that causes surprise to me how you Americans accept such bullshit.

        • Yeah, an earlier draft was better, but since you can only recover damages for KNOWINGLY false claims, and there are no statutory damages, it allows large-scale bogus claims. Truly, though, if it allowed damages for recklessly false or negligent claims, and had statutory damages, that would pretty much fix it. The procedure outlined in the law is actually pretty good. The content goes right back up if the person who posted it says it's not infringing. It's just the lack of any penalty for reckless claims

      • Re: (Score:2)

        by Rich0 ( 548339 )

        The problem is that DMCA doesn't effectively provide penalties for filing bogus notifications.

        It does require the complainant to make a statement under penalty of perjury. In theory false takedowns could be pursued in court.

        The real problem here is automated takedowns. How can you have a computer system make a statement on your behalf under penalty of perjury? It would be like sending a computer to testify on your behalf in court.

        • > In theory false takedowns could be pursued in court.

          The statute specifically says that if someone KNOWINGLY misrepresents tge facts in a DMCA notice, they can be sued for actual damages. In contrast , someone who NEGLIGENTLY infringes can be sued for statutory damages. Knowingly is a much huger standard than recklessly or negligently. If Google can prove that Warner Brothers KNOWS a notice they are sending is bogus, Google can sue for their actual costs, about $5. That's in the DMCA law , and that'

        • The DMCA requires that the person complaining have a copyright, or is authorized by a copyright author. This post is copyright by me, in the US and most of the civilized world, and so I can accuse any music site of infringing on it without committing perjury. (I believe that, if you can show that I knew the music sited didn't infringe on this post, you can sue me for actual damages, which are likely to be trivial.)

          As far as computer systems, I don't think the law recognizes autonomous machine action.

  • Optimizing the process (Score:3)

    by mdenham ( 747985 ) on Wednesday August 20, 2014 @11:51AM (#47713203)

    Just take down everything permanently because it'll eventually infringe another corporati--excuse me, "non-human person"'s copyright in the future anyway.

    • Re: (Score:2)

      by fnj ( 64210 )

      Just take down everything permanently because it'll eventually infringe another corporati--excuse me, "non-human person"'s copyright in the future anyway.

      I prefer the term "inhuman person" myself.

  • Well... (Score:5, Interesting)

    by Anonymous Coward on Wednesday August 20, 2014 @11:53AM (#47713229)

    "...parties are examining what they can do to optimize the process."

    Well, you could start by requiring that the entire notice be filed under penalty of perjury, not just the part that says you are who you claim to be.
    You could also start by requiring that the notice provide *evidence* (sufficient to sustain a claim of copyright infringement in a court of law) of the claimed infringement.

    Failure to do *either* or *both* of these is just going to result in the request rate increasing to the point where it is impossible for even a fully-automated system to handle them on the receiving end, because there's currently no downside to sending a bogus DMCA takedown notice.

    • I'm hoping that Google has no automation for takedowns, and only one person responsible for doing it, and that that person is handling requests on a first come, first serve basis.
  • These great numbers allow them to complain to our government corporate puppets that the DMCA is broken and needs "fixing", ie, make all search engines and service providers liable for infringement, thus requiring full keyword filtering, isoHunt style.
    • Or, they throw up a red flag that the system is being abused, giving our elected representatives cause to take away this special gift that has been given to the copyright lobby.

  • Why is this Google's problem? (Score:5, Insightful)

    by Last_Available_Usern ( 756093 ) on Wednesday August 20, 2014 @12:06PM (#47713359)
    They aren't hosting the content, they're merely making pointers to it. Isn't this an issue that should be handled by the company hosting/managing the web content? I'm surprised Google is getting involved in this at all and it makes me wonder what their motivations for doing so are, given the obvious administrative burden this is imposing.

    • Re:Why is this Google's problem? (Score:5)

      by jones_supa ( 887896 ) on Wednesday August 20, 2014 @12:17PM (#47713453)
      What if I erected a sign with the text "Small Cock -->" next to you? You maybe wouldn't want me to have the sign there. At that point would it be OK to you if I just said "I ain't hosting the penis, I am just making pointers to it"?
      • Actually, it would be more accurate if someone else was holding the sign and Google was just pointing it out. In this case I still say the guy holding the actual sign is the one you go after.

      • That's no where near the same. In your example, I'd have to be 1 in 10 billion people in a hidden room (the cloud), and for anyone to see the sign they'd have to explicitly search me out. Then upon emerging as a possibility for person search, your sign is only one of 10 other signs, several of which will be, "Big Penis," and "Huge Penis," because that's the honest to goodness truth, I tells ya. Then maybe they click on a link where I have a graphical depiction of the member in question.

        So it's completely

      • That analogy doesn't work. For it to work, lots of other people have to have put up signs pointing to you saying "Small cock." Google just provides a list saying who has the most such signs pointed at them. i.e. Google is just providing a numerically sorted statement of fact, not making up new signs. Removing an entry from Google doesn't change the fact that all those other such pointers out there in the web continue to exist and continue to say you have a small penis. Indeed, once you've used Google f

      • What if I erected a sign with the text "Small Cock -->" next to you? You maybe wouldn't want me to have the sign there. At that point would it be OK to you if I just said "I ain't hosting the penis, I am just making pointers to it"?

        Assuming you are not violating zoning laws, I see no problem with this. It is an issue of free speech. I think it sucks and I definitely would not like it, but I would support your right to point out my small cock.

    • Re: (Score:3)

      by tepples ( 727027 )
      What Google gets if it follows the OCILLA procedure set forth in 17 USC 512 is a safe harbor defense to accusations of contributory or vicarious infringement.

    • US courts have ruled that linking to infringing content constitutes contributory infringement. I'm not sure how "direct" the link has to be. Google search links to torrent sites which contain pointers to infringing content, for example, so that's apparently okay.

  • Time For Cynicism (Score:4, Insightful)

    by organgtool ( 966989 ) on Wednesday August 20, 2014 @12:07PM (#47713363)
    The rightsholders have claimed copyright on birdsong, a public transmission of the space shuttle launch, and many other claims of complete nonsense, proving that their algorithms are way too aggressive in flagging videos and that they can't even be bothered to review the "infringing" material before issuing a takedown notice. So who wants to bet that the legislative resolution to this issue has nothing to do with harsher penalties for fraudulent requests and everything to do with harsher penalties for "pirates" who happened to have a radio or television playing in the background when they caught something unusual on video?

  • Math wrong by one order of magnitude ... (Score:5, Informative)

    by Anonymous Coward on Wednesday August 20, 2014 @12:10PM (#47713385)

    86400 seconds in a day, 1 million takedown notice per day --> 1 notice every 0.0864 s, so 86ms

    Seriously, how hard is it ?

    • what if we say "every business day", so divide your number by 3. HA!

    • Re: (Score:2)

      by mcrbids ( 148650 )

      I told you at least a hundred million times that math is just too hard to do!

    • Re: (Score:3)

      by Rashdot ( 845549 )

      Oh good. I was thinking how hard that job must be, but every 0.0864 seconds is a lot easier then every 0.00864.

    • For anyone who wants to learn (eg: editors)

      Take downs per second:
      takedowns a day / seconds in a day
      1000000 / 86400 = 11.57407407407407

      For seconds:
      1 second / takedowns a second
      1 / 11.57407407407407 = 0.0864
      or
      seconds in a day / takedowns per day
      86400 / 1000000 = 0.0864

  • 1kk or 10kk (Score:1)

    by Anonymous Coward

    Doesn't 8ms add up to 10 million links rather than 1 million?
    24 * 60 * 60 * 1000 = 86,400,000ms
    86,400,000ms / 8ms = 10,800,000

    Eh...

  • Very, very easy to fix (Score:3)

    by Trailer Trash ( 60756 ) on Wednesday August 20, 2014 @12:18PM (#47713461) Homepage

    All takedowns have to be sworn under penalty of perjury. Next time google gets one that points to a page with no infringement (just happened) (just happened again) (oops, and again, okay, I'll stop counting now) whoever sent it needs to be prosecuted for perjury. The infringement notice bots would be shut down in 10 minutes when those behind them are suddenly facing prosecution.

    As I've said time and again: we don't need a new law - we need to enforce what we've got.

    • Re: (Score:3)

      by CPIMatt ( 206195 )

      The problem is that there is no penalty in the law for a mistaken copyright take-down. What we need is for every mistaken take-down, an automatic $1000 fine.

      -Matt

      • Re: (Score:2)

        by fnj ( 64210 )

        Automatic shit is very facile to propose, but it doesn't work. Since the Bartley-Fox law went into effect in 1975 in Massachusetts, anyone convicted of carrying an unlicensed firearm faces a mandatory one year jail term, no appeal, no parole. The problem is, said carriers have a better chance of being struck by lightning than serving that sentence. Massachusetts cities are of course filled to the gills with unlicensed firearms; they are used criminally every night; a fair percentage of perps are eventually

    • All takedowns have to be sworn under penalty of perjury. Next time google gets one that points to a page with no infringement (just happened) (just happened again) (oops, and again, okay, I'll stop counting now) whoever sent it needs to be prosecuted for perjury. The infringement notice bots would be shut down in 10 minutes when those behind them are suddenly facing prosecution.

      As I've said time and again: we don't need a new law - we need to enforce what we've got.

      They're under penalty of perjury, but only that the submitter is acting in good faith that they actually own the copyright to the content. If there's a link to a torrent of Cap_America_Winter_Soldier.mp4, and Disney files a takedown request based on the filename, they're fine, even if it turns out that the file is actually four public domain images of a baseball hat, a map of the US, a tree with snow on it, and a GI, all on a loop for two hours.

  • Google should fill... (Score:1)

    by Anonymous Coward

    Google should fill a request to be forgotten...

  • What do they consider a take down request? (Score:3)

    by ZiakII ( 829432 ) on Wednesday August 20, 2014 @12:36PM (#47713653)
    What do they consider a take down request though? For the website I administer for my job I just submitted 60 take down requests using the Google webmaster tools this morning as our Chat Form Page got incorrectly indexed (robots.txt error on our part) and we want our customers to go through the contact reason page first.

    These are legal requests but are they counting these in that number?

  • Optimize the process? (Score:3)

    by jenningsthecat ( 1525947 ) on Wednesday August 20, 2014 @12:37PM (#47713669)

    ...both copyright holders, Internet service providers, and other parties are examining what they can do to optimize the process.

    The solution is not to optimize the process. The solution is to scrap the process, and the DMCA along with it.

  • Competitive Sabotage? (Score:3)

    by GameboyRMH ( 1153867 ) <gameboyrmh@@@gmail...com> on Wednesday August 20, 2014 @12:43PM (#47713725) Journal

    You have to wonder if competing search engines could be spurring these claims somehow, culling such vast numbers of pages from Google's index is a great way to degrade the usefulness of Google's search...

    • Re: (Score:2)

      by ron_ivi ( 607351 )

      Might not be competing search engines.

      Could just be a SEO-trick where a company hires a SEO optimization company to remove the content of competitors.

  • First hand experience (Score:5, Interesting)

    by Charliemopps ( 1157495 ) on Wednesday August 20, 2014 @01:07PM (#47713893)

    I used to work in a department that handled DMCA notices on the consumer side. (They were complaining that our customers were hosting the content) The vast majority of these complaints were fraudulent. The problem is that the media companies hire other companies to monitor for infringement and send take-down notices. I suspect they pay per notice sent and they are getting swindled. Some were so bad, we literally blacklisted their domain so they'd stop sending us complaints. They'd send take down notices for people that weren't even in our IP block. They were just sending nonsense and collecting money from the content provider. This likely also where the content providers get their insane numbers about the amount of money they are losing.

    • And it's going to continue until there are real repercussions to sending invalid dmca take-down notices. Right now I can issue dmca take-down for every video ever added to youtube. What's going to stop me?

      • What's going to stop you from taking everything down? The fact that nobody will take you seriously. The DMCA process isn't mandatory to follow. What it does is protect the host from liability.

        Somebody sends a DMCA request. The recipient can do a takedown and then the recipient has done its legal duty, and has no liability from the alleged copyright holder. The recipient can pass this along to whoever put up the allegedly infringing content, and allow a counterclaim, and then the recipient is complet

  • ... they know acting like they are will achieve their objective to stop the nonsense.

    The best way to get regulators off your back is to over comply.

    That way, the regulators take the fall.

Slashdot Top Deals

Whoever dies with the most toys wins.

Close