Study Shows How Notice-And-Takedown Reduces Transaction Costs In Making Works Legally Available
from the so-of-course-the-entertainment-industry-wants-to-kill-it dept
We've written a few times in the past about research done by Paul Heald on copyright and its impact on the availability of certain content. He's recently published an interesting new study on how the DMCA's notice-and-takedown regime facilitates making content available by decreasing transaction costs among parties. As we've discussed at length, the entertainment industry's main focus in the next round of copyright reform is to wipe out the notice-and-takedown provisions of the DMCA. The legacy recording and movie industries want everyone else to act as copyright cops, and hate the idea that notice-and-takedown puts the initial burden on themselves as copyright holders.However, Heald's research looks at music on YouTube and concludes that the notice-and-takedown system has actually enabled much greater authorized availability of music, by reducing transaction costs. The idea is pretty straightforward. Without a notice-and-takedown provision, someone who wants to post music to YouTube needs to go out and seek a license. Of course, getting permission from all the various rightsholders is frequently impossible. The transaction costs of getting permission make it such that it's way too high. Yet, with notice-and-takedown, the person can upload the content without permission, and then the copyright holder is given the option of what to do with it. On YouTube, that includes the option of monetizing it, thus "authorizing" the use. That creates a natural experiment for Heald to explore, in which he can see how much content is "authorized" thanks to such a setup. And the result, not surprisingly, is that this system has enabled much greater authorized (and monetized) access to music than an alternative, high transaction cost system, under which uploaders must first seek out permission to upload everything.
In fact, the analysis shows a tremendous number of popular music hits from the US from 1930 to 1960 are available in what's likely an authorized (i.e., monetized) fashion, even thought nearly all of it was almost certainly uploaded by those without permission. Under the system that the RIAA and MPAA would like, this would be next to impossible. Instead, they'd want to negotiate deals first, making it nearly impossible for such works to be available, and meaning that both the availability and monetization of those works wouldn't be happening. As Heald concludes:
Congress should resist calls to dismantle platforms like YouTube which take advantage of current limits on secondary liability to create a marketplace that radically reduces the high cost of negotiating over rights to music and visual content. The access YouTube provides to valuable cultural products is far from perfect, but it provides a partial solution to the problem of disappearing works, at least in the music context. In any event, no new legislative initiative should proceed in the absence of concrete data testing the claim of copyright owners that their proposals make works more, rather than less, available to the public.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: content id, copyright, dmca, efficiency, markets, notice and takedown, paul heald, transaction costs
Reader Comments
Subscribe: RSS
View by: Time | Thread
[ link to this | view in chronology ]
Still not the best solution
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Correct me if I'm wrong...
Would those be music hits from before 1972? Oh! So they are. Now I wonder why they're not being challenged...
[ link to this | view in chronology ]
[ link to this | view in chronology ]
I think you are wrong, they want to destroy the Internet because it allows people to self publish, which destroys their business models. Further it is putting huge pressure on timed and regional releases, and regional distributors, which the publishers have manipulated to maximize their profits.
[ link to this | view in chronology ]
This is still not a reasonable system, due to the arbitary nature of decisions about 'infringing' content. It allows amateurs to gamble on a video staying up, and the studios to carry on with their expensive licensing arrangements. It does however make it very difficult for low budget professionals to publish works, as they cannot afford the license negotiations or the loss of revenue if the video is taken down or monetized by someone else.
One of the major flaws of the system is that it does not allow for fair use properly, by making the defense of fair use too expensive, which hurts in particular documentary video making, and videos of real life, like dancing babies, or other interesting or amusing things that people capture on video.
[ link to this | view in chronology ]
Re:
I think that's partly true. I also think that the apparent lack of (preferably severe - my opinion) consequences to copyright holders bogus take-down notices and baseless litigation are factors, together with similar lack of consequences to web sites for cooperating with bogus take-down notices are not helping. Sadly, that seems to point in the direction of courts not applying the law as written.
Do we need new law, new precedent, or maybe some of both? That is a question I don't have an answer for.
[ link to this | view in chronology ]
Re: Re:
Technically, filing a false or bogus DMCA claim is supposed to be treated as perjury, a crime that carries a jail sentence to it. The problem is, an already weak law has been weakened even more over time, such that the idea of anyone but an individual filing false claims against content owned by the big companies, for whatever reason, actually being punished for it... yeah, not going to happen.
And since there's no punishment, there's no need for accuracy, so they just submit claims on everything and anything.
Introduce potential jail time to the mix though, and you can be sure that they'd be a lot more careful of submitting claims, and actually check before just hitting submit.
[ link to this | view in chronology ]
Re: Re: Re:
I think it's only perjury if they knowingly file a false claim. As it's written now, they can just say "oops" and they're off the hook. Except there is no hook because perjury is a crime, so a District Attorney / US Attorney would have to file charges, and as you say that isn't going to happen.
[ link to this | view in chronology ]
Re: Re: Re: enforcing Jail time ...
Enforcement of this penalty would replace the current client base with one willing and able to pay for upgrades to their designated stay. The added benefit would be more openings for new law and MBA graduates.
[ link to this | view in chronology ]
They already a spending as little money as possible on their failure of a program to stop piracy by flooding 3rd parties, not actually hosting the content, with wild demands that things be magically removed.
They are fighting so very hard to keep time from moving forward and hold onto the glory days of the past, convinced that there is no money in digital, completely ignoring the record profits coming in from digital sales.
I mean we totally need to change the laws to be more unbalanced to protect these corporations and open up more abuses of the public, who cares if they are the largest ignored stakeholder in all of this... someone might hear a birdsong that a computer thinks sounds like a real song and not get them paid!!!!! Babies might dance to the barely audible music in the background, that someone will figure out how to turn into an MP3 and use to blow up the planet!!!
We totally need to nuke the net from orbit and go back the the gramophone and allow corporations to hold back all of society.
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re: Re:
[ link to this | view in chronology ]
Property that can walk
Freedmen's Bureau Records: L. P. Dangerfield to [unknown], July 15 1865
Rock Spring
Dr Sir
At a meeting of the citizens of the county on the last court day they passed a resolution that no one should hire the servant of another without the written permission of his master and as I had done so on the verbal statement of your servant that such permission had been granted but without having any written [added: permission] from you I at the request of the Boy write that he may get your consent to hire to me if you dont want him yourself. I have a great deal of work to be done and am in very great need of a hand being myself very weak handed. The neighbors are very much down on me for hiring him under the circumstances but I assure you I did it in good faith not intending anything wrong towards anyone.
Very Respectfully Your Friend
L. P. Dangerfield
[ link to this | view in chronology ]
Forgot something
I think you meant "legal availability"...
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re: Re:
moving infringing marerial on service providers. This is patently untrue as Masnick well knows.
[ link to this | view in chronology ]
Re: Re: Re:
To a third party, they have no way of knowing if a given file is infringing or not, as who put it up and under what circumstances can totally change the legal status of it.
[ link to this | view in chronology ]
Re: Re: Re:
moving infringing marerial on service providers
If this is not what rightsholders believe then they should stop acting like it.
And for fuck's sake, learn to spellcheck. You don't need more help making you look obtuse and dunderheaded.
[ link to this | view in chronology ]
Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re:
I submit that neither of the above scenarios make any sense.
[ link to this | view in chronology ]
Re: Re:
[ link to this | view in chronology ]
Re: Re: Re:
Though I suspect that was meant as sarcasm, yes actually, it is their problem, other people/services/companies are not beholden to help them beyond what the law requires, no matter how much some of them might wish it otherwise.
As for the 'numerous instances of obvious infringement', file a DMCA claim on it, and if it really is that obvious, then odds are good it will be pulled, but again, they are not beholden to rights owners to proactively police that stuff, and in fact shouldn't be expected to, as what is infringing, and what is not, is not nearly as black and white as you seem to think it is.
[ link to this | view in chronology ]
Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re:
Are you daft, or is there something else that motivates you to look for excuses?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re:
Only the publisher know what licenses they have granted. Unless they put both the content, and license up in s searchable database, ISPs and other service providers do not know what has been licensed, and what is infringing. Even the labels have been known to take down content that they put up, and they are the ones with the necessary information to know what is licensed and what is infringing.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
If they do not know what licenses have been granted, then how is anybody else meant to know what is or is not licensed.... magic?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
Thus demonstrating conclusively why it's MORONIC to argue that service providers should magically know.
and even if it did it has no realistic and reliable way of identifying whether or not a specific copy is the subject of a lawful license
And yet... you think service providers should be liable if they don't magically figure out the very thing you're now claiming the rightsholders themselves can't figure out.
Are you serious?
Add to this a bit or realism...thousands of copies can be uploaded in seconds, and yet you seem to expect that a rights holder must spend every waking moment tracking them down.
Clearly not the case, since so many use a variety of automated services. Besides, in the alternative you have a much more ridiculous situation in which the service provider -- with no insight or knowledge into the legality of a work -- would need to do the same.
This is an impossible task
And you want to make it more impossible by dumping it on a party with LESS knowledge into what's legal. Again, are you mad?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
"My goodness. Look at all the time being wasted by innovators having to send out notices." "What, you the rights holder actually want the name of my subscriber, and are trying to get their name using long accepted legal procedures?" "Give me a break. Why should I have to cancel the account of someone who repeatedly breaks the law?"
The DMCA incorporates an exception to some tort law doctrines regarding secondary liability. Without the exception service providers would have much more skin in the game and would have to pay much more attention to what is transpiring at their site. It hardly seems fair to lay everything at the feet of rights holders when historical data demonstrates an imbalance of burdens that arises because of the exception.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
If you're going to demand someone's name to drag through the courts and wear them down through legal attrition you'd damn well better make sure you've got the right person, not some unlucky sod who doesn't even have a computer.
Unfortunately rightsholders have frequently been proven to be incapable of the trust and responsibility to possess that sort of power.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Moving the goalposts I see.
You claimed, directly, that it was *impossible for rightsholders* to determine what was licensed and what was not -- and that was a reason that the burden should be on unrelated service providers.
I pointed out that your own argument nullified your own argument.
And you ignored it and went with snark.
Because, once again, we've called you out for your ridiculous claims. And, once again, you will never admit that you made a ridiculous argument that proved your own argument was wrong.
The DMCA incorporates an exception to some tort law doctrines regarding secondary liability. Without the exception service providers would have much more skin in the game and would have to pay much more attention to what is transpiring at their site.
This is undoubtedly true. And, as we've pointed out repeatedly (though you ignore it) the end result of having to "pay much more attention" would be massive censorship that served to shut off all sorts of legitimate forms of communication, and which would likely lead to many service providers no longer allowing open communications because the threat of liability is too high.
You may like the idea of fucking over the public so that broadcasters go back to having a controlled platform, but I do not. I find it disgusting.
It hardly seems fair to lay everything at the feet of rights holders when historical data demonstrates an imbalance of burdens that arises because of the exception.
False equivalency. There is no "imbalance of burdens" at all. The burden is *properly placed* on the *only party* in a position to know what is, and what is not, infringing. As you yourself has demonstrated, the possibility that the service provider could have such knowledge is "impossible."
And yet you wish to put that burden on them, clearly knowing that the risk of liability will almost immediately turn open platforms like YouTube into closed platforms like Hulu. To people like you, this is a "benefit." To the public it's a disaster. But years watching you work has made it clear that your entire focus is on helping the powers that be fuck over the public.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
But you are asking for reallocation of burdens that are unachievable.
That service providers should remove or disable instances of infringement when pointed out to them has already been established. They already do this right now.
You are now asking for the service providers to determine whether infringement occurred without giving them enough information to make such decision.
The rights holder MUST provide the service providers with the information needed. It can be a blacklist system, like we have now with notice and take-down or a whitelist system where the rights holders identify the authorized instances. Either way, providing the information HAS to come from the rights holders and cannot be reallocated to the service providers because they have no way of knowing what is authorized or not.
This is why just about every response to you has pointed out that what you really what is a "magic" system where the service providers do all the work and the rights holder reap all the benefits without lifting a finger. That's crazy talk.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
If you have a real, solid argument for why and how service provers should be "allocated" any of the work of determining what is or is not infringing, I'm totally willing to consider it.
As of now, I've yet to hear such an argument. You haven't even made one yourself. So, what is it?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Viacom vs. YouTube
Are you blind? Did you somehow overlook Viacom vs. YouTube, in which Google presented evidence that Viacom authorized the posting of numerous "pirated(-looking) videos" on YouTube as a form of viral marketing?
Somehow, your "daft" is my reality. Sorry, dude.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re:
That's asking for the impossible. Copyright infringement isn't recognizable by the content alone, it's an actionable offense based on the circumstances of usage.
If the rights holders want a system that comes close to "magically" removing all instances of infringement, then the rights holders need to provide a whitelist, since only they actually know what's authorized or not.
Moving copyright back to "opt-in", where registration is required, would be a fine place to start creating such a whitelist.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re:
Guess that it is a bit difficult for them to use keywords as a first cut for possible problems...words like "hack", "keygen", "serial numbers", etc., etc.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re:
Why is that a problem, exactly? It's how copyright worked prior to 1976.
Assuming this was a requirement, no doubt service providers would fall back on "But how are we supposed to know if that specific file uploaded all over the place on our site is an infringing copy? No, you have to check each out and then let us know before we are willing to take any action."
The rights holders would certainly have to be proactive in maintaining and updating such a whitelist. But again, I'm not sure why that's such a burden, since the rights holders ARE the one benefiting from an exclusive right to monetize the works.
Guess that it is a bit difficult for them to use keywords as a first cut for possible problems...words like "hack", "keygen", "serial numbers", etc., etc.
Because, once again, depending on what and how the "hack", "keygen" or "serial number" is used determines infringement, not just whether it exists or not.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
/s
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
This sounds like a generalization for which there is no supporting evidence.
File sharing is not illegal. I assume you have fallen into the bad habit of referring to copyright infringement as file sharing - the two are not the same. To wit, the computer you used to post your comment(s) on this website did so by sharing files with the server operated by the ISP you do business with. Said ISP server then shared those files with multiple routers/switches en route to the TD servers. File sharing is ubiquitous on the internet, but not necessarily in the way you think.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
You yourself noted above that even the rights-holder can at times not know what's licensed or not, so how in the world are third parties supposed to know, magic?
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
1) Not. Their. Problem. The law as it currently stands does not force one industry to spend time and money working for another industry as their enforcers, if the labels/studios want infringing works to be pulled, it's up to them to find it and file a claim on it.
2) And how are they supposed to know it's there without the copyright owner pointing it out to them?
Smaller service providers are likely dealing with tens, if not hundreds of thousands of files on their services, not only are they not obligated to search through all that on a regular basis, if would be a herculean task to ask them to do so for all the potential infringing files, especially given a point you seem determined to refuse to admit, that of the fact that they cannot know what is and is not authorized.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Your pov, I see it as being relevant.
"File sharing" is inappropriate when referring to copyright infringement.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
Funny thing to say, given you seem to be wanting to lay all the work on service providers to act as copyright cops, instead of the copyright owners.
However, realistic examples, I can certainly do that. From the following examples, tell me which are A) Infringing, or B) Not infringing.
As you seem to fall under the 'Infringing files are obvious and easy to spot' category, and moreover expect third parties to decide the legality of files based on what limited information they have, you will only be given what information they'd have, the name and file type.
1. A text file named 'Harry Potter collection'
2. A video file named 'Night of the living dead'
3. A set of archive(zip, rar, whatever) files named 'Horror movies', 'Actions movies', 'Sci-fi movies' and 'Fantasy movies'
4. An mp3 file with the name of a big name musician and one of their latest songs.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
It is obvious is not a valid answer, as it assumes a level of information you are unwilling to provide.
[ link to this | view in chronology ]
Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re:
[ link to this | view in chronology ]
Re: Re: Re: Re: Re:
Where are these instances you claim exist? I am not aware of any infringing material located on the servers run by the business which provides me internet access. Perhaps you have conflated service provider with website.
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Cases like these entered the debate concerning Section 512 of the DMCA. The benefits of promoting the growth of the web were a major factor in crafting a statutory accommodation exempting liability so long as certain criteria were observed. While the current version of the statute is satisfactory in many instances, it does impose a burden on rights holders that in many instances is arguably unfair.
To my knowledge no one is advocating the wholesale elimination of rights holders burdens. What they are advocating are possible changes to Section 512 that may better align the respective roles of rights holders and service providers.
[ link to this | view in chronology ]
Re:
What is even more unfair is trying to make third parties liable for massive damages because they did not take action because of lack of knowledge. There is no reliable way for an algorithm to recognize content without having a validated sample available to compare against. Further there is no way to know what are authorized up loads when there is no practical way to know what permissions the up-loader has. For intance The MPAA and RIAA members made significant use of MegaUpload including paying for premium accounts.
[ link to this | view in chronology ]
Re: Re:
This said, the DMCA does take this concept into account, and in its drafting sought to ameliorate somewhat its impact on online service providers. Apparently some here believe this mainstay of tort law simply has no place online because, well, it is online and that makes it special. Most assuredly it does not because tort law reflects the actions of people and their relationships, and these actions and relationships hold true even in the digital world.
[ link to this | view in chronology ]
Re: Re: Re:
No, it is you who is wanting to make special provisions because it is online. Automobile manufactures don't get blamed for illegal uses of their vehicles, so why should ISP's be blamed for how their customers use the tools they provide?
Most assuredly it does not because tort law reflects the actions of people and their relationships, and these actions and relationships hold true even in the digital world.
And so does copyright law, online and offline. It's also dependent upon the actions of people and their relationships to determine if infringement occurred. It cannot be determined by the existence of the content alone.
What you are asking for would be tantamount to asking Ford Motor Company to vet each and every car buyer prior to purchase so that Ford could be held liable if someone speeds with one of their cars. That is not what we do in the offline world - we hold the individual responsible for their own actions. It's you who seems to think the internet requires a different set of rules where you get blame the entity with the largest pocketbook, instead if those who are actually responsible.
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]