[HN Gopher] The epidemic of fraudulent DMCA takedowns
___________________________________________________________________
The epidemic of fraudulent DMCA takedowns
Author : colinprince
Score : 115 points
Date : 2024-02-17 17:45 UTC (5 hours ago)
(HTM) web link (taxpolicy.org.uk)
(TXT) w3m dump (taxpolicy.org.uk)
| ralmidani wrote:
| This epidemic will continue to spread unless the burden of proof
| is placed on the party filing the complaint, along with having
| penalties imposed for filing bogus complaints.
| LegitShady wrote:
| the problem is the scale of the content is so big that no
| company is actually willing to have humans review all the
| claims.
| tokai wrote:
| That should really be their problem to solve.
| realusername wrote:
| If they can't even notice the abuse, does the abuse even
| matter? I think that's the core problem here.
| dangrossman wrote:
| Imagine you're Disney and you want to crack down on
| merchandise infringing on your IP. You type [site:etsy.com
| "minnie mouse"] into Google and get 3.8 million results.
| There's another million results for "iron man", 2.1 million
| for "avengers", 6.8 million for "star wars", etc. That's
| just the likely infringement on a single marketplace. I
| don't think this is a scale too small to notice. How many
| people would Disney have to employ to review every listing
| before making a claim, and continue to do so as fast as the
| world population can make additional unlicensed copies?
| realusername wrote:
| Maybe they just don't need to simply? I understand they
| would like as much control as possible but that's
| impossible with the abundance of choices and places
| nowadays and I firmly think that those automated blind
| takedowns are doing more harm than good.
|
| I don't think total control is possible in such an
| environment, DMCA or not.
| devmor wrote:
| With all due respect, that sounds like a problem for
| Disney.
|
| I do not see why a corporations scale of intellectual
| property protection should have its burden foisted upon
| everyone else.
|
| It makes to sense to provide a modicum of protection that
| can apply to individuals, but just as individuals can't
| require a policeman to stand guard outside their home
| 24/7, corporations must hire security guards if they want
| extra protection from the undesirable elements their
| commercial activity attracts.
| j-bos wrote:
| Seems like a problem for automation to solve, automation
| takes existing jobs, workers transition servicing in niche's
| that are currently lacking, like all forms of moderation and
| review. Oversimplified of course.
| paulgb wrote:
| That would be the ideal, but a secondary deterrent is the
| Streisand effect.
|
| If I had the time, I would love to aggregate listings from
| Lumen that use sites like uniontimes and thetribunepost to
| instigate false takedowns, and list them all so that they would
| come up in a Google search showing that someone had
| (fraudulently) tried to have the stories removed. I'm not sure
| how much of a deterrent it would be, but I imagine that sort of
| thing would at least prevent people from recommending the fake-
| takedown approach to others.
| vladmdgolam wrote:
| Similar thing happens in the music world, where Kanye West
| recently interpolated a Donna Summer's track for the Vultures 1
| album, and Apple Music & Spotify immediately deleted the track,
| even before any official process. The lyrics and melody were
| different, so it's not even an interpolation in a way, and all
| that did not matter.
| supriyo-biswas wrote:
| From a technical perspective, Google could compare whois domain
| registration details and snapshots at the Internet Archive with
| the allegedly infringing URL.
|
| There could be mandatory identity verification for filing DMCA
| complaints as well, but I'm not sure whether that would gel well
| with the current regime which penalizes not acting on
| infringement notices. Further, as someone whose application
| assets were copied without my permission by small time firms and
| had to go the DMCA route, I'd probably want to avoid giving out
| my ID to every hosting company under the sun.
| SllX wrote:
| I agree with you in sentiment but this part:
|
| > From a technical perspective, Google could compare whois
| domain registration details and snapshots at the Internet
| Archive with the allegedly infringing URL.
|
| No to Google abusing the Internet Archive's servers. If they
| want to host a mirror of their own and hit that, fine, but you
| know what they could be using instead? Their very own Google
| Cache.
| thayne wrote:
| > Their very own Google Cache.
|
| You mean the thing they recently announced they were shutting
| down, and suggest you use the Internet archive instead?
|
| [1]: https://arstechnica.com/gadgets/2024/02/google-search-
| kills-... [2]: https://searchengineland.com/google-search-
| officially-retire... [3]:
| https://news.ycombinator.com/item?id=39234246
| SllX wrote:
| The very same.
|
| My response here applies as well:
| https://news.ycombinator.com/item?id=39413967
| knodi123 wrote:
| > Their very own Google Cache
|
| They permanently disabled that recently.
| SllX wrote:
| I know. That's why I said it. Abusing a non-profit's
| services for a useful function when you had your own to
| make use of? That's what we call a dick move. Although this
| is a hypothetical dick move. Killing off Google Cache was a
| dick move and wasn't hypothetical though.
| tagyro wrote:
| could, would, should ...who is Google? What's their address?
| Hell, what's their email address? Has anyone been able to write
| an email to Google? (</s>)
| kevingadd wrote:
| "We have robust tools and processes in place to fight fraudulent
| takedown attempts, and we use a combination of automated and
| human review to detect signals of abuse"
|
| Google lying to the public like this should be considered false
| advertising. They don't even require takedowns to be fully filled
| out.
| olliej wrote:
| No, google is doing what it can within the extremely tight
| confines of what the DMCA allows.
|
| The DMCA essentially prohibits companies from applying any
| judgement as to the validity of the complaint, so all google
| can do is try to identify fraudulent filers, but if a new
| entity makes a complaint it doesn't seem like there's a huge
| amount of signal available at that point to indicate whether
| that entity is legit.
|
| The issue with the DMCA is it says something like "if you try
| to decide whether or not claims are valid, you also accept
| liability for infringement", which means there's near unbounded
| cost if you try to reject claims because the claims themselves
| are invalid, e.g. the DMCA "allows" a corporation to do that
| filtering but only in a way that no one would ever do.
|
| Imagine you were running a forum, and you got a DMCA takedown
| notice for something someone posted. Let's say it's clearly
| obvious that the claim is invalid, and you reply with "hell no,
| that's clearly a BS claim". Later on someone posts something
| that is obviously copyright infringing. As I understood what
| people were complaining about (some 20 years ago sorry),
| because you rejected the earlier takedown you may now be liable
| for that infringing post. Hence, the only safe option available
| is to always start the two week takedown.
|
| For many of these fraudulent claims that's the goal - they just
| want to censor the content. It doesn't matter if they ignore
| any future contact attempts, it doesn't matter if the person
| who got hit provides you evidence the claim is false, the DMCA
| basically requires that you keep the content down for two weeks
| to allow the original claimant to respond (which in these
| fraudulent cases they won't).
|
| Google obviously does stop bring stuff back early now if it
| gets major publicity, but my guess is they have more lawyers
| than most companies can afford, and that they believe that if
| they were to go to court they could say they did their best to
| comply with the DMCA but the pressure (often incl. politicians)
| made them feel it was in the public interest or something to
| revert the takedown or whatever, and that that argument would
| hold up in court.
|
| The problem is not Google here, the problem is a trivially
| abusable law that has no penalties for fraud.
| kevingadd wrote:
| I'm not talking about "is this fake", I'm saying you can send
| Google a "dmca takedown" that doesn't provide all the
| information required by the DMCA, and they will still take
| the content down.
| olliej wrote:
| Ah, womp womp, you'd think that at least would be correct.
|
| My guess is they got enough shitty/badly done forms from
| real entities filing real reports and got threatened with
| lawsuits when they were rejected instead of the filers just
| going "oh, we'll fill out what we agreed to fill out"
|
| Sigh.
| Nextgrid wrote:
| This is just a symptom of the wider problem out there that the
| world is wildly unprepared to deal with cases of industrial-scale
| fraud.
|
| The DMCA takedown process assumes that nobody would make false
| declarations under the assumption that there would be
| consequences to doing so.
|
| For a while, this worked - the consequences (or the mere
| possibility of them) - kept everyone more or less honest.
|
| The problem nowadays is that bad actors are starting to see
| through the illusion.
|
| This is not limited to DMCA takedowns either - corporation-
| against-consumer fraud has also been normalized.
| olliej wrote:
| There was never a penalty for incorrect DMCA claims. Unless you
| mean perjury?
|
| Perjury does not have any meaningful penalty because you can
| always claim "I honestly believed this claim was correct, but
| now I see how I was wrong" - e.g the numerous takedowns against
| material that is clearly fair use. Then even the blatantly
| incorrect claims have to be responded to with an immediate
| takedown because the DMCA not only doesn't allow the processing
| organization to confirm the claimant owns the IP, or represents
| the firm that owns the IP, it doesn't even provide the option
| to confirm the claimant is a real legal entity that exists. If
| the claimant doesn't even legally exist, then even if they are
| committing perjury there's no one to bring the claim against.
|
| It's also again kind of moot, because making these complaints
| realistically requires getting a lawyer, but the DMCA again
| does not provide any mechanism to recover legal fees arising
| from incorrect or fraudulent claims.
|
| When the DMCA was passed everyone with half a brain was saying
| that it was clearly trivial to abuse.
| masklinn wrote:
| The DMCA also does not put the accuracy of the claim itself
| under penalty of perjury, so false claims have no penalty
| _whatsoever_ under the DMCA, not even perjury, only
| performing a takedown while not authorised by the
| rightsholder.
| miki123211 wrote:
| This also assumes that the claimant is even under US
| jurisdiction.
|
| Let's say you are affiliated with the <insert dictatorship
| here> government and want to censor memes critical of <insert
| dictator>. You can pretend you're John Appleseed, living at
| 123 Church Street in Maine, and that the memes are copies of
| your own work, posted without attribution.
|
| If anybody dares to file a counterclaim, they have to provide
| all their personal details. If they post <insert dictator
| here> memes, it's likely they have some connection to your
| country, either by straight up living there or having family
| members. You can exploit those connections and do whatever
| dictators like to do with people who like to post memes
| critical of them.
|
| If you find an actual American with no connection to your
| country who isn't afraid to disclose their personal details
| to you, they can sue. They can probably even win, possibly in
| a default judgment, as you're not from the US and therefore
| unlikely to show up in an American court. That, however,
| doesn't matter much, as there's no way the US can do anything
| to you. Even if they figure out who you are, they can't
| forcibly take money out of a bank account in <insert
| dictatorship here>.
| Gibbon1 wrote:
| I think the answer is a fraudulent DMCA complaints are wire
| fraud. With the addition that you can be charged if there is
| a pattern of recklessness.
| ajross wrote:
| > the DMCA not only doesn't allow the processing organization
| to confirm the claimant owns the IP
|
| This isn't a correct way of understanding things. The DMCA is
| silent about what the hosting provider is "allowed" to do. It
| simply says that _IF_ you take it down following a DMCA claim
| _THEN_ you cannot be held liable for any infringement that
| occurred while you hosted it.
|
| If the claimant doesn't hold the copyright, then there can't
| have been any such infringement anyway. But yes: if you do
| your own due diligence about the IP ownership and get it
| wrong, then you can be found liable for copyright
| infringement. So most providers are trigger happy out of an
| abundance of caution.
| masklinn wrote:
| > The DMCA takedown process assumes that nobody would make
| false declarations under the assumption that there would be
| consequences to doing so.
|
| That is dubious, given the DMCA does not define any
| consequences for false declarations. It does mention penalty of
| perjury but that's not even for accuracy of claim, it's _only_
| for the complaining party being "authorized to act on behalf
| of the owner of an exclusive right that is allegedly
| infringed."
|
| DMCA abuse took off almost immediately, here's a 2005 study:
|
| > Examining the characteristics of the targets of the notices--
| the alleged infringers--we found that 41% of all Google notice
| targets can be classed as competitors of the complainants. Fig.
| ES-4. This is particularly significant for Google 512(d)
| complaints regarding links in the index, where 55% of all
| notices relate to competitors. A significant percentage of the
| 512(c) and (d) notices sent to Google 21%--target hobbyists,
| critics, and educational users.
| kelseyfrog wrote:
| What's an acceptable rate of false positives and false negatives?
| pixl97 wrote:
| In a typical civil court cases, what is the ratio of wins
| (including settlements) to losses by the plaintiff?
| 1letterunixname wrote:
| No shit. DMCA is an authoritarian disaster and a gift to rent-
| seeking corporations and to criminals. Remember the little-to-no
| evidence legitimized extortion prosecution campaigns of terror.
| sonar_un wrote:
| Classical musicians fight a similar battle every day with Sony,
| UMG, and Warner filing copyright on every single piece of
| classical music posted online.
| a_wild_dandan wrote:
| Could the artists band together, collect evidence, and create a
| class action lawsuit? Or is the juice not worth the squeeze?
| sonar_un wrote:
| Classical musicians are too poor in general to fight this.
| asdadsdad wrote:
| https://news.ycombinator.com/item?id=37034931
| bluish29 wrote:
| The problem is with the economy of the DMCA take-down process. It
| is almost free to file on a scale, very high cost for content
| host to review on a scale and for alleged infringer it you will
| need a DMCA counter which no incentive to accept because it is a
| risk. Then the only option is to sue which mean high cost against
| sometimes a Cooperation with army of lawyers which is not worth
| it even if you are right. There is fundamental imbalance in the
| process.
| ronsor wrote:
| Courts require filing fees. Why can't websites require them for
| DMCA?
| _Algernon_ wrote:
| The solution here is to abolish copyright. It is a law intended
| for _publishers_ not _individuals_ , which results in this kind
| of abusive exploitation when applied to individuals who have
| little means to defend themselves.
|
| No, really. We live in a post-scarcity world when it comes to
| information, but artificially keep pretending it is scarce. Of
| course that will cause problems.
|
| Keeping copyright in a post-scarcity world is like forcing every
| individual to jump through financial regulation hoops intended
| for corporations each time they buy their morning coffee.
| chrisjj wrote:
| Publisher individuals are a thing.
| _Algernon_ wrote:
| They weren't when copyright was created which is the point.
| Hence the metaphor for financial regulation and coffee.
| olliej wrote:
| "The solution here is to abolish copyright"
|
| If you abolish copyright, then open source licenses cease
| mattering.
|
| The GPL only works because a corporation can only use GPL
| software if it agrees to the license, but if there is no
| copyright then there is nothing for the license to gate access
| to. Alternatively if you take the more recent argument that
| "the GPL is a contract not a license", then the contract also
| does not work because a core component of contract law is that
| a contract must have some kind of reasonable exchange and if
| there's no copyright the source software has no value (IANAL
| but the gist as I understand it is you can't have a contract
| that exchanges something for nothing).
|
| If you think software is bad now, wait until every piece of
| software, even the non-game ones, basically operates in the
| cloud so you never get direct access to any of the actual
| software because if you can download and run it locally you can
| give it to anyone.
|
| You also ensure that authors of music and books cannot make
| money, because they have no ownership of the work they produce.
|
| What your copyright-free world does is give a huge advantage to
| corporations with huge amounts of cash over pretty much
| everyone else.
| _Algernon_ wrote:
| >What your copyright-free world does is give a huge advantage
| to corporations with huge amounts of cash over pretty much
| everyone else.
|
| _At worst_ maintaining the status quo, in other words.
| mistrial9 wrote:
| friend, I deeply disagree with your conclusion and your
| reasoning. The current situation is in need of guidance and new
| rules certainly. However the strident "don't do this to us"
| approach is wrong-headed from my own experiences and knowledge.
| There are a dozen major economic systems across the globe with
| billions of people involved, and those system do not agree with
| each other on this topic. There is no single answer for all
| people, all markets and at all times. In the USA and Western
| Europe, copyright is a foundation of individual authors and
| also companies, and builds an economic system of exchange that
| has worked in many cases, and for long periods of time. Abuses
| by powerful entities are not the reason to throw the entire
| system out, with insults and broad generalizations IMO.
| Robin_Message wrote:
| I'm confused by this argument.
|
| Copyright exists in order that creative works are sold for more
| than their marginal cost of reproduction, and the profits from
| that are directed towards the author. Having the exclusive
| right to produce the copies, copyright, enabled creativity to
| be rewarded.
|
| Making distribution cheaper and easier makes copyright perhaps
| make _more_ sense, not less, but didn 't change the underlying
| logic of trying needing to sell creative works for more than
| their marginal reproduction cost, even if you figure that cost
| is zero.
|
| Except for scientific publishing. Burn that all down. Because
| the authors aren't getting any money (and actually want
| distribution, not payment) so copyright of scientific works
| isn't achieving anything positive for society at all.
| tomxor wrote:
| > but Google rather brilliantly accepts takedown notices without
| checking if the person filing it exists
|
| These days, if was going to publish anything critical on my own
| website about any company, I'd put it on NearlyFreeSpeech.
|
| > while we aren't lawyers, neither are we idiots. We can tell the
| difference between people harassing our members via the DMCA and
| cases where our service is genuinely being misused, and we can
| adjust our attitude accordingly.
|
| https://www.nearlyfreespeech.net/about/faq#DMCAResponse
| olliej wrote:
| I love presenting this as being google facilitating the fraud.
|
| They're literally doing what the DMCA requires, and everyone at
| the time it was passed said would be abused: Companies are
| essentially required to assume claims are legitimate or they
| become liable for infringement, add to that no penalties for
| incorrect claims, no cost for making the complaint, the
| requirement for the claimed material to be kept offline for two
| weeks iirc even if the recipient proves ownership, and this is
| expected.
|
| There are many reasons to not like google, but this is not
| googles fault.
|
| This is the DMCA operating exactly as it was designed and
| written.
|
| If people want to stop DMCA abuse, the solution is:
| * Incorrect claims have an increasing penalty for every incorrect
| claim involving the same companies or people. e.g. overlapping
| boards, overlapping lawyers, overlapping claimants, etc.
| Incorrect meaning it's fair use, etc * False claims -
| the claimant does not own the IP, the IP is out of copyright or
| similar - are treated as criminal perjury, and in addition to the
| penalties for incorrect claims the fair and reasonable copyright
| penalties of 10k for individuals or 100k for corporations for
| every blocked viewing based on average views leading up to the
| DMCA hit * The person making the claim must cover the
| full legal costs of the recipient in the even the claim is
| rejected (currently recipients can't afford a lawyer to respond
| to even the most blatant fraud) * If the recipient can
| provide - under the standard perjury penalties that currently
| don't appear to apply to claimants - demonstration that they are
| the owner/creator of the IP the content goes back up, and the
| legal representatives of the parties can be put in contact with
| each other (either to argue over ownership, or to arrange for the
| entity making the claim to pay the legal fees and penalties)
| * The processor handling DMCA requests can require that the
| entity making DMCA claims provide full contact information for
| the IP holder as well as the claiming entity as well (because
| large companies outsource IP enforcement to legal firms with no
| assets to recover, again why G and co should be able to require
| ahead of time funds to cover recipient legal expenses). The
| processor is permitted to ignore claims from any organization
| where it cannot confirm all contact and identity information, and
| it is permitted to charge a reasonable fee for those identity and
| background checks. Recall that the DMCA takedown system is a
| shortcut and convenience to circumvent the legal system, so a
| company can also just engage the legal system normally if they
| don't want these fees. * If a company or organization
| that makes repeated incorrect claims, the minimum period before a
| takedown increases to two weeks. The processor can require the
| claimant demonstrate that they have the funds to cover the
| recipient's legal fees in the event the claim is rejected and if
| any recipients of invalid claims are unable to have their legal
| fees and the penalties repaid in a timely manner, the processor
| can require an escrow to cover those fees ahead of time and not
| service any request without such - per claim - escrow set up.
| * The various restrictions on repeat claimants, etc apply to the
| actual holder of the IP, the legal firms representing them, and
| the individual legal representatives signing documents. e.g. you
| can't be a hansmaier type firm where the same lawyers keep re-
| incorporating as new legal entities to circumvent the
| restriction, and similarly if you are a legal representative that
| keeps filing incorrect or fraudulent paperwork then if you move
| to a new company you would poison that companies claims as well.
|
| Essentially, if you want to keep the takedown mechanism of the
| DMCA there have to be penalties if you abuse it, and corporations
| processing those claims need to be able to ignore claimants that
| abuse the takedown process.
| indymike wrote:
| The solution to this problem is to simply require a small, say
| $50-$100 deposit that is paid to the host (e.g. youtube) the
| site/account/content owner (e.g. youtuber) if they file a
| counterclaim (covering up to some reasonable number of urls) and
| the DMCA claimant does not file a Federal lawsuit. If there is no
| counterclaim, then the deposit is refunded. If the claimant
| defends their copyright with a Federal lawsuit, the deposit is
| refunded.
|
| This makes takedown mills and fake take downs economically
| unfeasible and is reasonable for claimants in that it awards some
| money for dealing with fraudulent and mistaken claims.
| blacksqr wrote:
| A simpler solution using existing infrastructure is simply to
| require a takedown notice to include the copyrighted work's
| registration information.
|
| It's not necessary to register a work with the US Copyright
| Office to have copyright protection, but registration is
| necessary to file an infringement claim in court.
|
| Registering takes time and money, and the registration database
| is public. Simply requiring a URL that points to the work's
| registration as part of the DMCA notice should take care of the
| problem.
| tagyro wrote:
| But then someone has to check if the URL matches the
| complaint, right? Or?
| blacksqr wrote:
| The content host has to do work to locate the accused
| infringing content and make it inaccessible regardless,
| checking a URL for a content match should be a minimal
| additional effort.
| indymike wrote:
| > registration is necessary to file an infringement claim in
| court.
|
| You can register post-infringement and sue anyway.
|
| > Simply requiring a URL that points to the work's
| registration as part of the DMCA notice should take care of
| the problem.
|
| I don't think an amateur photographer should have to pay a
| registration fee on every photo.
| blacksqr wrote:
| > You can register post-infringement and sue anyway.
|
| Never said otherwise. But if the copyright holder is
| serious about stopping the infringement, then registration
| will have to be done if the infringer files a counterclaim,
| in order to go to the next step.
|
| > I don't think an amateur photographer should have to pay
| a registration fee on every photo.
|
| They wouldn't, just the ones that are infringed.
| dangrossman wrote:
| Google says there are 7 million search results for "Star Wars"
| on Etsy.com, an online marketplace for handmade goods. They're
| likely all infringing on Disney IP, as anyone big enough to get
| a license to make official Star Wars merchandise isn't going to
| qualify as an artisan handmade crafter to sell on Etsy.
|
| To perform a crackdown on those 7 million listings, you're
| proposing that Disney would need to make a $700 million dollar
| deposit with Etsy and wait for hopefully most of it to trickle
| back in. If the infringer is somewhere on the other side of the
| globe that Disney can't easily reach legally, they can't get
| their deposit back until they spend thousands filing a federal
| lawsuit that'll go unanswered, so the deposit is more than
| forfeit in many of these cases. That's for a single company to
| defend a single franchise on a single website.
|
| I don't think this is a "simple" solution.
| grecy wrote:
| So you're saying one of the richest corporations on the
| planet can't afford to put down $700 million to guarantee
| their claims of infringement are actually accurate?
|
| Good, don't make the claim. Maybe multi-billion dollar
| companies shouldn't be enforcing copyright in such a
| draconian way in the first place.
| indymike wrote:
| > They're likely all infringing on Disney IP,
|
| First, let's not use IP here. Let's be specific and say
| copyright since that is what DMCA covers.
|
| > Etsy.com
|
| This is an awful example as etsy would be the company that
| the DMCA takedown would be against. Etsy certainly is worth
| suing or profiting from selling unlicensed goods.
|
| > $700 million dollar deposit
|
| The amount of the deposit you use is the high end of the
| range. Even so, you didn't account that each deposit covers
| up to 100 URLs, so the deposit would be $70M which I think
| the mouse can afford, easily.
|
| > they can't get their deposit back until they spend
| thousands filing a federal lawsuit that'll go unanswered,
|
| The deposit would go back if the takedown does not get a
| counterclaim. Legitimate claims would result in the deposit
| being returned. Not every claim will have a counterclaim. If
| there is a counterclaim, the next step is a Federal lawsuit
| anyway. And if they don't answer you get a default judgement
| that can be collected in many countries around the world, and
| even on contingency.
|
| > I don't think this is a "simple" solution.
|
| You gave very little consideration to the people that are
| badly affected by fake counterclaims, and are assuming that
| every actor is a bad faith factory.
| michaelmrose wrote:
| Google's number of results page is a complete lie told by
| Google there never were 7 million results to start with.
| There are probably only single digit thousands of listing
| with 99% of listings from less than 100 people. Sue them and
| forward the legal papers to Etsy.
___________________________________________________________________
(page generated 2024-02-17 23:00 UTC)