There is a new legal decision by the Sixth Circuit Court of Appeals in the United States that brings together cheerleading uniforms and 3D printing. It isn’t a game changer – the decision doesn’t fundamentally rewrite legal rules – but it does shed some light on what is likely to be one of the thorniest issues around 3D printing and intellectual property law: a copyright concept known as severability. The question addressed by the court was seemingly straightforward: are the decorations on cheerleading uniforms a critical part of their functionality? The way the court found an answer to that question will help us understand what types of 3D printed things are – and are not – protected by copyright.
I clicked on the link but now am thinking I don’t want to read all of this: what’s the short version?
Severability is the test that US courts use to figure out which parts of things that are both functional and decorative can be protected by copyright (if any). There is now a new test to determine this in part of the United States. The test itself (arguably the tenth test currently being used) doesn’t bring a lot more clarity to the issue, but is an excuse to walk through how all of this works. Most of this blog post just explains how the court applied the test in this one case.
Still with me? Let’s go!
Background Point 1: Useful vs. Creative
As explained in the Introduction to Rights and Protections post, there is something of a divide between objects that are eligible for copyright protection and objects that are eligible for patent protection. Creative objects can get copyright protection while functional objects can get patent protection. In theory these two categories are mutually exclusive. In practice, some objects can be a little bit of both. Courts use “severability” to try and determine if the mixed elements of a single object can be separated out and protected individually.
Oh, and as a general matter clothing and uniforms are considered a functional object – they keep you warm and not naked.
Background Point 2: Severability
While a lamp is a useful article, the leg sculpture that forms the base of the lamp can be severed (both physically and conceptually) from the lamp’s utilitarian function and given independent copyright protection. Image from flickr user Spider.Dog.
Severability is key to understanding copyright protection in objects that mix functional with non-functional items. As a general rule, functional items do not get copyright protection. However, if a court can sever non-functional elements from functional elements, those non-functional elements can be protected by copyright. This severability can be “physical” (because the parts can be physically separated from each other) or “conceptual” (because the parts can be conceived of separately from each other). All of this stuff is covered in more detail in the What’s the Deal with Copyright and 3D Printing? whitepaper.
Enough – What is This Case About?
Cheerleader uniforms. A company named Varsity Spirit designs and sells cheerleader uniforms. Their uniforms include colors, stripes, chevrons, and other patterns. Star Athletica makes cheerleader uniforms with similar combinations of colors, stripes, chevrons, and other patterns. Varsity accused Star of copying their uniforms and brought a copyright suit against Star. Star responded that it could not infringe on Varisty’s copyright because the patterns on the uniform where not eligible for copyright protection.
Essentially, Star said that Varsity’s patterns were functional parts of the cheerleader uniform. They helped mark the uniform as a cheerleader uniform as opposed to, say, just a pleated skirt and tank top.
That’s what makes this a severability case. Varsity says that the patterns are severable from the utilitarian uniforms and therefore should be eligible for copyright protection. Star says that the patterns are a core part of the uniforms which cannot be severed – and therefore cannot be protected by copyright.
How Did the Court Work Through This?
Severability is easy(ish) to explain but hard to apply in specific circumstances. After noting that there are at least nine (listed on pages 17-19 of the opinion) different tests of severability in use, the court decided to come up with its own 5(ish) part test.
Part 1: Is the design in question something that could be protected by copyright?
Yes. In this case the court decided that the work Varsity registered with the Copyright Office was a “pictorial, graphic, or sculptural work, ” which is the type of thing protectable by copyright.
Part 2: Is the design of a useful article?
Again yes. Uniforms are a useful article. Their use is to “cover the body, wick away moisture, and withstand the rigors of athletic movements.”
Part 3: What are the useful aspects of the article? Importantly, “conveying information” doesn’t count as a use.
The court decided that the designs on the uniforms were primary intended to convey to people that it was a cheerleader uniform. Unfortunately for Star, in this court’s opinion “conveying information” like that doesn’t qualify as a useful purpose. I think it is fair to say that this is not a universally held opinion among courts.
Part 4: Can the viewer of the design identify the artistic features separately from the useful features?
The court found that it could. The designs on the uniforms did nothing to help cover the body, wick away moisture, or withstand the rigors of athletic movement. As such, they were unrelated to the core functionality of the uniform. This is where the answer to part 3, and how you define the functional elements of the uniform at the start of the analysis, becomes important. Once the court decided that helping to identify a uniform as a cheerleader uniform didn’t count as a useful function, it was hard for them to find any other way.
Part 5: Can the designs exist independently of the useful object?
This question is really the core of the analysis and, truth be told, the first four parts don’t help answer it very well. The court walked through a bunch of sub-questions to try and answer this part. Ultimately they decided that the same designs could exist independently of the uniform and could, for example, be transferred onto things like jackets or pants (or a framed wall hanging).
As a result of all of this, the court found that the designs were severable from the uniforms and therefore wereprotectable by copyright.
What Does All of This Have to do with 3D Printing Again?
Although this case was about uniforms, it isn’t hard to imagine applying the same analysis to 3D printed dog cupsor tea light holders. Understanding if a 3D printed object is protected by copyright becomes an important question if someone is copying it without permission. If the object is protected by copyright, that copying may be infringement. If the object isn’t protected by copyright, there may not be very much the designer can do. When you are thinking about building on some existing 3D model (or someone has built upon or copied your model) it is important to have a sense of how it may and may not be protected by copyright.
And if you think that this all feels a bit too complicated you are not alone. In a dissent, Justice McKeague reflected upon the multitude of ways that someone could think about severability and declared that “the law in this area is a mess.” He then went on – with some reason – to call on either the Supreme Court or Congress to straighten all of this out. While severability remains important, until either the Supreme Court or Congress clarifies some things, actually applying it is going to remain a challenge.
Super Copyright + Administrative Law Nerd Bonus
The court declined to grant the Copyright Office’s decision to register the works Chevron deference as to whether or not the designs were copyrightable. That’s one more example of how the Copyright Office isn’t really an administrative agency with actual rulemaking authority. If this paragraph is gobbledygook to you, you can safely ignore it and go on with your life.
Shapeways designers create a mindboggling array of objects every day. This diversity fosters a vibrant Shapeways community and makes Shapeways shops exciting places to explore. It also means that models on Shapeways can potentially involve a number of different types of intellectual property and other rights.
Sometimes this rich diversity of rights can be a bit confusing and overwhelming. In an attempt to bring a bit of order to the rights chaos, today we’re kicking off a series of blog posts that try to briefly describe a handful of types of rights that might touch upon models on Shapeways. Today’s post will cover the high level basics. Future posts – they’ll be coming up every fortnight or so – will dive a bit deeper into issues raised by these rights. We’ll try to link to all of them from this post in the future so you can find them all in one place.
(If you want a slightly more in depth – but still written for designers who are not lawyers – explanation for how all of these things fit together, you might want to check out thesethreewhitepapers).
((This series is for educational purposes only and is not legal advice. This area of law tends to be maddeningly fact-specific, so if you have a specific question it is always wise to talk to a lawyer who understands your circumstances.))
A drawing of a pig dancing with children is protectable by copyright.
If you are familiar with one kind of intellectual property right, it is probably copyright. Copyright is designed to be an incentive to produce artistic, creative things. Here on Shapeways that can mean things like sculptures, figurines, and jewelry. Importantly, copyright generally does not protect functional items like mechanical parts or brackets and connectors (those are the domain of patents, described below).
You do not need to apply for a copyright in order to have copyright protection (although there are advantages to registering – something that you can do here.). As long as the thing you have created is of a type that is eligible for copyright protection, it is protected by copyright from the moment it exists. Copyright protection will generally last for the life of the author plus 70 years after her death. Copyright also allows for independent creation – if two people come up with identical models without copying each other, each one gets their own copyright protection for their model.
Copyright mostly governs the making of copies of a protected item (that’s where the “copy” in copyright comes from). Traditionally this focused mostly on ‘literal’ copying, like making an exact copy of an entire book or origami crane skeleton. However, copyright also governs the creation of ‘derivative’ works. These aren’t necessarily exactly copies, but rather are copies that are derived from some protected work. A classic example of this is creating a “copy” of a book by turning it into a movie. There are important exceptions to these general rules (fair use probably being the best known) that we will cover in future blog posts.
A dynamo is the kind of invention that is protectable by patent.
Patent is designed to be an incentive to produce “functional” innovations. Contrast this with copyright – those mechanical parts, brackets, and connectors that I said were not eligible for copyright protection are the types of things eligible for patent protection.
Also in contrast with copyright, you do not get a patent automatically just for creating something that is functional and original. You need to apply for a patent in order to get patent protection. If you get it, that patent protection generally lasts for 20 years. Unlike copyright, patent law does not allow for independent creation. That means that once you have a patent, anyone who practices that patent will infringe upon your patent even if they didn’t know about it beforehand.
Generally speaking, someone infringes on your patent by “practicing it” – that is, doing what the patent describes – without your permission.
You can trust Aspinall to sell a product that is “absolutely non-poisonous” so look for their trademark on the bottle.
Copyright and patent are both awarded as incentives to create and share innovation. While trademark is often lumped in with copyright and patent as “intellectual property,” its purpose is quite different. Trademark is all about consumer protection. When you see a trademark on a product, that trademark means that the trademark owner stands behind the product.
As a result of this different purpose, trademark rules can feel very different from copyright and patent rules. In copyright and patent, infringement usually comes in the form of copying or using the protected creation. In trademark, infringement is usually judged by a “likelihood of confusion” standard: is it likely that a given use of a trademark would confuse a consumer into thinking that the trademark owner is standing behind the use? Does the use make it seem like the object is from or endorsed by the trademark holder? If so, the use may be infringing on the trademark.
Right of Publicity
TR might not like it, but as a public, newsworthy figure Puck is not violating his right of publicity even when they use his image to sell magazines.
Right of publicity is also a bit out of the intellectual property mainstream, but it is worth mentioning here. Right of publicity is all about being able to control how your image is used commercially. In the US it is based in the right of privacy. Generally speaking, if you are leveraging someone’s identity for commercial gain you may be infringing on their rights of publicity.
There is at least one noteworthy exception to this general rule. In the United States we don’t want the right of publicity to smother the type of free expression protected by the First Amendment. As a result, the more you are using someone’s likeness to comment on or criticize them – and this is especially true if that someone is a public figure – the less likely you are to be infringing on their right of publicity.
Just an Introduction
This post is designed to be an introduction to some of these concepts and, as a result, just scratches their surface. It leaves a lot of things out that may or may not be relevant to a specific case. Some of those things will be addressed in posts further into this series.
If there are issues that you would like addressed in future posts let me know in the comments or on twitter@MWeinberg2D. We can’t address specific legal disputes, but we can explain how some of these pieces fit together.
David Byrne’s op-ed in the Times on Sunday got me thinking about the importance of information sharing in online music services. In that spirit, I decide to throw up this unpublished whitepaper from 2011-12 on ways forward for online music licensing. Central to the idea is complete transparency on who is consuming what. It provided a way for unlicensed services to become licensed and to minimize the economic value of unauthorized music distribution. It certainly has its flaws (it fails to appreciate the importance that labels put on pricing different songs differently, which may or may not be a legitimate thing to consider in this context) but I’ve always kind of been partial to it. Remember that at the time this was written Spotify had pretty much just launched in the US and its licensing situation was from from clear. Anyway, for posterity’s sake here is the text of the whitepaper. It never had a title.
The fundamental idea is that instead of subscribing to a service, you subscribe to music. You can take your music subscription to any service that offers you music in an interesting ways. That way, instead of competing on the licensing deals that services can cut, services compete on being good at helping you to experience and discover music.
The music industry provided many
people their first opportunity to consider how the internet would impact
existing industries. Napster became
synonymous with the first widely recognized internet “crisis.” The RIAA, the lobbying organization representing
record labels, has been a driving force behind both strengthening digital
copyright laws and enforcing those laws in court. However, even as the internet has moved from
novel to commonplace, after almost fifteen years the music industry has not yet
fully come to terms with the realities of digital distribution.
Although the music industry was one
of the first industries to be impacted by widespread consumer access to the
internet, it has lagged behind others in finding ways to adapt to the new
connected digital landscape. In fact,
“avoid being like the music industry” has become something of a guiding
principle for other industries facing similar digital disruption.
Period of Stability
That is not to say that the music
has not taken important steps towards engaging with the internet. Compared to the apocalyptic rhetoric and
wall-to-wall litigation of the early 2000s, the current online music landscape
appears to be relatively stable. It is
easy and convenient to pay for music files from services such as iTunes and
Amazon. Google, Amazon, and Apple (among
others) are beginning to offer consumers ways to stream their music collection
on demand from an internet connected device.
Grooveshark, Turntable.fm, and Spotify give consumers on-demand access
to a wider range of tracks. Pandora
creates customized radio stations for each individual user.
This stability does mask problems,
however. Compared to many internet
sectors, the online music landscape is not diverse. Large companies dominate the market for
purchasing music downloads. There appear to be few, if any, opportunities for
startups to enter the pure download market.
Even Apple’s “revolutionary” iTunes store simply met a need that had
been clearly expressed by consumers for years.
Only in the music industry are such late-to-market developments like
selling single tracks and albums digitally online considered milestones.
Streaming is similarly hobbled. Large incumbents are just beginning to offer
consumers the option of streaming the consumers’ own music collection
online. Smaller companies trying to
offer similar services have been smothered by lawsuits. Services such as Grooveshark and
Turntable.fm, unlikely to be able to negotiate full licenses with major record
labels, rely on a combination of relatively untested legal theories and limited
licenses to protect themselves from liability.
Pandora struggled unprofitably for years, fighting with record labels
for an economically viable licensing agreement.
Even Spotify, the darling of the online European music world, took two
years to negotiate its stripped down entry into the United States market. Even now the “successes” in this area are
trailed by complaints from many sides regarding any number of issues.
There is a reason that there are so
few truly successful startups in the digital music world: licensing. For the vast majority of music startups, it
is simply impossible to negotiate an economically viable music license. For this reason, veteran venture capitalist
Tony Conrad has compared investing in digital music startups to fighting the
The music industry has been unable
to tap into the startup culture that is creating thousands of new killer apps
and exciting services.
These apps and services would increase
demand, interest, and engagement in music, hopefully to the benefit of all
involved. The music industry is largely
left out of the energy, excitement, and engagement of the digital world because
it has not found a way to come to terms with the services that make that world
Why Can’t the Industry Support
There is no inherent reason that the
music industry should be unable to support innovation and change. Many parts of the industry are in an almost
constant state of change: new genres and artists are continually appearing to
upend existing mainstays. Major labels
have entire departments devoted to constantly discovering new content. It is clear that the music industry does not
hate change and innovation in and of itself.
However, the music industry is
having a hard time coming to terms with the best way to embrace the internet. Faced
with diminishing profits, record labels are constantly in search of a new
digital revenue stream. Unfortunately,
labels also expect this stream to be able to restore revenues to industry high
water marks. If a new opportunity does
not offer a clear path towards hundreds of millions of dollars in revenues,
many labels do not think that it is worth their time. Similarly, labels are worried that striking a
smaller deal with a service that proves to be very popular and profitable will
cause them to lose out on the next big thing.
Essentially the record labels, like many industries, is haunted by the
fear of waking up one morning to realize
that they have traded analog dollars for digital pennies.
This caution has made record labels
reluctant to incubate new music startups that could compete with their existing
(but declining) offline revenue streams.
They insist that startups pay steep advances, guarantee minimums, and
accept licenses that force them to operate under slim margins and restricted
feature sets. Ambiguity surrounding which types of licenses
are required for various types of offerings further complicates the terrain.
The shortsightedness of this
strategy becomes more apparent by the day.
The record labels are rapidly moving towards a point where they no
longer have to worry about trading analog dollars for digital pennies because
only analog pennies remain. Furthermore,
the labels are beholden to a handful of large online companies (especially
Apple) that were able to meet the licensing terms in the past. Instead of a diverse landscape of companies
offering a multitude of music related services online, most digital music income
is reliant on Apple, and to a lesser extent Amazon.
Until the record labels find a way
to make it easy for music startups to innovate and grow, while at the same time
paying for the use of recorded music, this pattern is likely to continue.
The absence of a straightforward,
easy to obtain blanket license for online music distribution prevents the music
industry from fully monetizing the strong desire of music fans to consume music
online. It also effectively prevents the
large number of application developers that want to distribute music online
from acquiring necessary rights (or even identifying which rights are
necessary). If those developers do
manage to acquire those rights, the absence of a reliable license creates a
looming threat that success will result in crippling licensing fees. This paper describes a workable framework for
a blanket license for online music.
design, this system allows all of the parties in the music industry to focus on
what they do best. Artists can focus on
making music. Labels can focus on
nurturing and promoting talent.
Collective rights organizations can focus on tracking, collecting, and
distributing money. Equally importantly,
it brings a new party into the industry: application developers. The license allows developers to focus on
designing new ways to get people interested in music (and interested in paying
for music) instead of negotiating complex licensing agreements. The public will be able to purchase a license
and be allowed to access music however they see fit.
Briefly, the proposal is for a
voluntary collective license. Users pay
a monthly fee to a digital content licensing organization. In return, they get a username and password
that they can bring to third party services to access music in any number of
ways: streaming, downloading, curated stations, etc. In exchange for reporting usage statistics to
the digital content licensing organization, any third party can get permission
to distribute music. However, third
party services do not receive a percentage of the license fee: they are
responsible for creating other revenue sources.
The digital content licensing organization then distributes the
collected fees to artists proportional to public consumption.
providing the public with all of the music they want, delivered in the manner
they prefer, while at the same time creating a dependable revenue stream for
artists, this license will serve as a way for the music industry as a whole to
embrace digital distribution. As an
additional bonus, it will avoid making the industry uncomfortably dependent on
one corporation or entity for its profits.
Nature of the License
license will be voluntary. This aspect
serves a number of important goals.
First, it is the only politically feasible way to create the
license. Rightly or wrongly, the digital
public does not trust the music industry.
Imposing a required fee on all Internet connections, or all digital
devices, or on anything else, will be seen as an attempt by a struggling
industry to tax innovation. This
perception will be compounded because, if a mandatory fee were to be imposed on
the public by the music industry, other industries (such as the news, movie,
and publishing industries) would quickly insist on a similar fee.
Second, the parties responsible for
collecting a mandatory fee are unlikely to enthusiastically participate. ISPs have expressed skepticism about
involvement in any sort of fee collection, be it voluntary or mandatory. Becoming the fee collectors for the music
industry, and essentially becoming the public face of a potentially disliked
program (if it is mandatory), is unlikely to appeal to any intermediary.
Third, a voluntary license provides
a check on the music industry. Customers
will feel that they have the power to opt out of the license if they are no
longer interested in acquiring new music.
Although such opting out is unlikely to occur in large numbers, simply
having the option to stop payment will go a long way towards public acceptance
of a license fee.
a voluntary license allows people who are not interested in music to avoid
paying for it. If a consumer is
interested in music they can pay for music.
If they are uninterested in music, they can avoid doing so. While the percentage of people who are
completely uninterested in music is relatively small, such people do
exist. Financially supporting musicians
cannot be a prerequisite for Internet access.
The music industry does not want to be regulated as a public good, and
therefore can not have the power to impose general taxes for support.
of Third Party Applications
One of the most important parts of
the collective license is that it will allow for the creation of a diverse
group of third party applications and stores dedicated to convincing people to
become more engaged with music. The
history of music-related online startups clearly illustrates that there is a
great deal of interest in creating new ways for fans to connect with music. However, most of those attempts have
foundered on the rocks of licensing agreements.
By removing delicate licensing negotiations from service creation, a
blanket license will quickly lead to scores of new music-based applications.
Although the license will allow
licensees (the public) to acquire music from wherever they please, it will not
automatically absolve third party services of liability for assisting in that
acquisition. Services will need to
comply with a handful of simple requirements in order to obtain their license
to distribute content to members of the public licensed to receive it.
The most important of these
requirements will be to report the activity of its users to the digital content
licensing organization. This reporting
will be done in an aggregated, anonomized way that will reflect which songs are
being accessed, but not in a way that would compromise the privacy of
individual users. The digital content
licensing organization would establish a standardized reporting process that
would allow services to automatically deliver usage data. This data would be used to calculate payments
to artists and rightsholders.
Another requirement would be that
users log in to use the authorized services.
When users pay for their license, they will receive login credentials
(i.e. a username and password). In order
to access online music services, the consumer will enter those
credentials. This validation process
could occur once (associating a license username and password with a specific
account on a specific service) or on an ongoing basis (using the license
username and password as the login credentials for a specific service). Existing fraud protections could largely
prevent a single account from being shared between large numbers of
Beyond these relatively simple and
easy to implement requirements, third party applications would be free to
distribute music however they see fit.
This would make it easy for existing black market applications to become
legitimate, licensed communities capable of attracting investors (if they were
so inclined). Applications could offer
track downloads, track streaming, curated radio stations, custom mix tape
downloads, or any other way they chose to package music. Because applications would not receive a
percentage of the license fee, it would be up to them to develop a profitable
of the License
The license must include both
downloads and streaming of tracks. While
there may have been a time when there were clear distinctions between downloading
and streaming music, that time has clearly past. Any number of services now allow consumers to
stream entire music collections on demand to connected devices. Although there are important distinctions
between downloading and streaming in the context of track ownership, innovative
services should have the ability to combine these offerings as freely as
Both downloads and streams should be
covered by the license. That does not
mean they should be compensated identically.
The compensation section (below) will address how best to accommodate
these differences when allocating payments to artists.
Point of Sale
potential place for consumers to purchase their license is with their primary
Internet connection as an addition to their monthly bill. After all, some sort of Internet connection
will be a prerequisite for using the license.
The license will be billed in monthly cycles. It will cover an entire household, just as
Internet service does today.
However, ISPs as a point of sale are
not ideal. In many instances, customers
have at least two ISPs – wireless and wired.
Being asked to purchase a license twice could lead to confusion.
the license could be offered directly to consumers by the digital content
licensing organization. Although
allowing consumers to purchase the license directly from the digital content
licensing organization might have a negative impact on the initial subscription
rate, in the medium and long term it could prove to be a more sustainable
model. It would also remove a burden
from ISPs and simplify the payment chain.
is reasonable to assume that the vast majority of users would migrate to
“legitimate” platforms once they are available.
This would occur both because new, attractive services would appear and
because many existing darknet services would incorporate the license into their
operation and “go legit.” Among other
things, this would allow these services to attract a broader, more lucrative
set of advertisers.
a result, most of the use data would come directly from the information that
apps reported to the digital content licensing organization. If this did not occur, the per-use counting
could be augmented with statistical sampling techniques from other sites. Of course, the digital content licensing
organization would also implement measures to combat clickfraud and other
attempts to artificially influence the data.
The digital content licensing
organization will allocate payment to rightsholders based on the percentage of
downloads and streams that each work received over the course of a period of
time (such as a month). Streams will be
weighted less than downloads, in a ratio to be determined in the future. The distinction between the two will be a simple,
straightforward rule such as “if the song can be accessed without an Internet
connection, it is a download.” The fixed
ratio model will allow third party services to offer the public downloads,
streams, or both, and to change the allocation according to market demand
without having to fear being crushed by new licensing fees.
The Digital Content Licensing
Organization: Rights Clearance and Division of Funds
digital content licensing organization must be an independent, transparent,
nonprofit organization. Its board would
be a mix of artists, labels, service providers, and others adequate to give all
parties confidence in the organization. It
will be responsible for creating and maintaining an authoritative database of
rightsholders, and for making that database accessible to the public. The organization will create open, standard
ways for third party applications to submit usage data. It will also collect funds from users and
distribute those funds to rightsholders.
In order to accomplish this task, the organization will retain a
percentage of the licensing fees to cover expenses. Finally, the organization would have the
power to set the license fee paid by the public.
While the blanket license solution
is an attractive one, implementing it will not be without challenges. None of these challenges are insurmountable,
but they must be dealt with in a realistic and transparent manner.
Creating the Index
The description of the solution
simply assumes that there is an accurate way to match songs used by services
with parties who should be compensated for that use. Although conceptually straightforward, today
this process is far from easy. In
addition to the fact that music tracks involve two distinct copyrights (one for
the underlying musical work and one for the actual performance), ownership for
each of those copyrights is often subdivided between numerous parties. Ownerships can be transferred in ways that
are poorly documented and hard to determine.
Companies with ownership interests can go out of business, and
individuals can die without their estates in order.
It is no understatement to describe
the creation of a universally inclusive, accurate index of music rights
ownership as a holy grail of online music. The fact that such an obviously useful tool
does not exist well over ten years into the online music story is a testament
to how hard it is to create.
That does not mean, however, that it
is impossible to create. The information
needed to populate the index exists today, scattered across various
organizations and companies.
Organizations like ASCAP and Harry Fox Agency have indexes that they use
to license works every day. Record
labels generally have the information necessary to license performances in
their catalog. The challenge is to find
a way to convince all of those organizations to share their information, and
then standardize and unify what they share.
Dealing with Non-Licensed Works
Without statutory backing, the
license must be a voluntary license.
That means that artists would be free to choose to join the license or
to not join the license. As a result, it
is possible that some artists would remain outside of the scope of the license.
Of course, this problem is not
limited to the license discussed in this paper.
Existing collective rights organizations, such as ASCAP and BMI, also
offer blanket licenses that are voluntary.
While some artists could theoretically decline to license their songs
through ASCAP and/or BMI, prowl the streets waiting to hear their songs
performed publicly, and then sue the establishment for copyright violation, in
practice this rarely happens.
is a simple reason for this – signing up with ASCAP is an easy way to guarantee
a revenue stream from public performances of a work. Instead of patrolling the world for copyright
violations, the artist can focus on making music. Also, the ubiquity of being “in” the license
means that licensed music is effectively cheaper than music outside of the
license. Once a business owner has paid
for an ASCAP license, any money paid to license other music is an additional
Ideally, the same dynamic would
occur with the license proposed in this paper.
The digital content licensing organization would control a large pot of
money, and be looking to distribute that money.
Most rightsholders would see the value in participating in that process
and focusing on music, instead of going it alone and patrolling the internet
Streaming/Downloading Price Differences
The fees paid to the digital content
licensing organization would be distributed proportionately to
rightsholders. It is legitimate to
consider how to take into account the real differences between streaming and
downloading when making this calculation.
As noted above, defining the
difference between downloading and streaming can be complex, but simple rules
of thumb could be applied. The key
distinction could turn on the accessibility of the file when the device is not
connected to the internet.
Alternatively, it could rest on what happens to the file if the user
stops using the service. True ownership
of a file that persists after service cancellation is more valuable than access
conditioned on paying monthly fees.
In any event, the ultimate question
is not necessarily how to define the distinction between streaming and
downloads, but rather how to set the compensation rations. How many streams of a song should count as
one download? This negotiation may be
somewhat easier than it has been in the past because the result will not result
in increased costs for users. Instead,
the outcome will only impact the percentage of the funds collected that will be
distributed to a specific rightsholder.
kind of sharing agreement where fans pay for music, artists get paid, and
innovators can innovate in a predictable licensing environment is an
intuitively attractive solution to many of the challenges in today’s music
industry. It is, unfortunately, by no
means an easily implemented one.
However, the goal of this whitepaper is to succinctly describe a
possible solution in hopes of giving all parties a single goal to work towards.
this is not the first time such a solution has been proposed, nor will it be
the last. Neither is it the only
possible solution. Artists are by their
nature creative, and the internet gives them a near limitless sandbox in which
to experiment with new business models.
New labels are forming with the goal of working with the opportunities
presented by the internet. Some innovators are finding ways to raise funds and
obtain licenses within the existing systems.
And, finally, there are people within major labels who are working hard
to orient their companies towards the new reality.
transition from analog to digital is never easy, and the music industry is no
different. It is perhaps the music
industry’s unique bad luck to find itself at the nexus of a history of sloppy
record keeping, a legal structure that grants powerful rights in a diffuse
group of people, and a product that was one of the first to be easily
distributed (both legally and illegally) online. Real change will require coming to terms with
all of that, and much more.
This proposal is not the first to try and address the
problems facing the music industry online and, regardless of the hopes of the
author, it is unlikely to be the last.
Similarly, the ideas expressed herein are build largely on the ideas of
others. For other thoughtful
examinations of the music licensing issue, you might consider starting with any
or all of the following:
 This paper is focused
primarily on the recorded music
industry. This is because recorded music
was traditionally the economic engine of the industry. A number of studies suggest that, while
revenue from recorded music is in decline, other revenue streams such as
touring are actually increasing and replacing that lost income. See, e.g. Michael Masnick & Michael
Ho, The Sky is Rising: A Detailed Look at
the State of the Entertainment Industry (2012), http://www.techdirt.com/skyisrising/.
However, although the decline of the recorded music industry is not synonymous
with the decline of the music industry as a whole, it is significant. The proposal made in this paper is an attempt
to find a sustainable way to monetize recorded music online.
 Experiments like OpenEMI
and Cantora Labs are certainly encouraging, but fall short of a full-scale
commitment by the industry. Furthermore, while musicians can (and are) free to
experiment online, doing so today generally means turning away from the label
system. Among other things, this can
make it harder for the artist to provide a centralized contact point for easy
The most surprising thing to me about working at Shapeways
so far is how people respond to copyright takedown notices. In fact, it is fundamentally changing the way
I think about online copyright debates.
Before I came to Shapeways I worked on online copyright
policy issues, so I thought I had a pretty good understanding of the copyright
notice and takedown process. A user
uploads a file to a website like Shapeways.
A brand sends Shapeways a notice claiming that the user is infringing on
their rights. Shapeways takes the file
down and tells the user.
In the classic model, this process gives the user a choice:
she can recognize that she is probably infringing and keep the file down, or
she can claim that the brand is wrong to order the file down - maybe because of
fair use, maybe because she has a license, maybe because the brand just picked
the wrong file by mistake - and order it back up. There is a lot of nuance lost in this model
(many people who should challenge takedowns are intimidated by the legal
process and don’t, for one), but it is still the classic textbook model for how
the process works.
While this model plays out every day, there is an additional
option that I didn’t fully appreciate: the user recognizes that they are
infringing but desperately wants to be put in contact with the brand so she can
get a license. She wants to find a way
to partner with the brand, sharing revenue and working together. In many cases, the takedown notice is a gift
because it is the first time she has been able to communicate with anyone from
the brand at all.
A Third Way
online copyright debate is often characterized by two archetypes: the unrepentant pirate and the stifled free speaker. The
unrepentant pirate works hard to undermine global creativity by making music,
movies, and everything else freely available online without regard for consequences. The stifled free speaker (or, if you prefer, proud mother sharing videos
of her child with her family) is unjustly silenced by overzealous copyright
There is a truth in each of these, while at the same time
each paints with too broad a bush. But focusing
on this dichotomy masks a third, growing category: the eager partner.
The eager partner does not lurk in the
shadows of the internet, motivated by a desire to make all copyrighted content
available for free. Neither is she
making expressive use of content in a way that does not require approval from
brands. Instead, the eager partner is consciously building on
existing IP – often IP that she has a deep emotional connection to – in a way
that probably does require some sort of license agreement.
She is not trying to hide from the brand. She does not think that the brand’s opinion
of what she is doing is irrelevant. The
eager partner wants more than anything else to become an official partner of the brand, and is more than happy to send the brand a percentage of
her revenue in exchange for doing so. She just doesn’t have a way to make that
partnership happen. (These
characteristics set her apart from the rich
trove of fan artists whose work is protected by fair use).
Potential Win Win With Partnerships and Licenses
Today, brands are failing to take advantage of these eager
partners. The eager partner is not
trying to avoid the brand – the eager partner wants to work with the brand. There are obvious benefits to this sort of
partnership. For the eager partner,
there is the ability to share – and sell – their creations with full approval
from the brand. The eager partner also
gets a deeper connection to the brand they love enough to use as a basis for
The brand benefits as well.
In the short term, the brand gets to avoid the awkward situation of
threatening a super fan with legal action.
The brand also gets the benefit of the eager partner’s market research
and testing. No matter how big a brand
is, it will never be able to imagine all of the products that might find a
market online. That means they are not
producing things that could be making them money. 3D printing enhances the cost of that
limitation because it even allows a market of 5 or 500 to be met profitably. In aggregate, these products can drive
significant licensing fees. If fans are
selling products, a brand will get a cut of each of those sales.
Trying Something New
This is a slightly different way of doing business, and it
won’t be as easy as waking up one day and deciding to do it. Embracing fans means giving up bit of
control. However many benefits that
loosening of control may bring, giving up a bit of control is always hard.
Traditionally, transaction costs have also been a barrier to
this sort of agreement. Dealing with
large numbers of fans – some of whom will not be bringing in large numbers of
dollars – can be logistically time consuming and expensive relative to the
Fortunately, neither of these challenges have to be deal
killers. I’ve argued
for a while that brands can set up straightforward licensing rules while
reserving the right to pull licenses for fan creations that make them feel
uncomfortable. That helps reduce
transaction costs by standardizing agreements and avoid awkward or
controversial creations by making it clear to eager partners what the rules
Oftentimes, giving up a bit of control is just what is
needed to reinvigorate and inspire a fan base.
Even a small amount of attention from a brand can go a long way in
strengthening the relationships with their fans. Also, there is no need to open the floodgates
immediately. Starting with a handful of
trusted fans can give brands the ability to test the waters and learn what is
Like many Americans, Adafruit decided to celebrate the recent 4th
of July with some patriotic music. They
scored a video celebrating the 4th of July and US manufacture of
Arduino with America the Beautiful, as performed by the United States Navy Band
(a work on the public domain – more on that below). Soon thereafter, they received a
note from YouTube telling them that someone had issued a copyright claim
against the video. After some outrage,
the claim was dropped. But what actually
happened here? And why did it happen?
The Song is in the
Let’s start with the song.
Musical copyright is super annoying (for a great dive into one small
part of music-related copyright – the ability for artists to reclaim rights
they gave to labels, check out this
whitepaper by my former PK colleague Jodie Griffin), but this instance is
fairly straightforward. A recorded song
has two copyrights: the copyright in the underlying work (the song as written)
and the copyright in the performance of that work. Let’s take them one at a time.
The underlying work is America the Beautiful. Wikipedia is as good a source as anything for
this, and it says
that the music for America the Beautiful was originally written in 1882, the
poem that became the lyrics was originally written in 1895, and the two were
combined in 1910. Using the copyright
domain calculator of your choice, we can be confident that works originally
published on those dates are in the public domain (I’m leaving out a possible newer
arrangement of the music for this analysis because no one has suggested it was
a factor and it complicates things).
Copyright one: check.
The performance is the recording of the actual band
performing the song. It is pretty likely
that this performance happened at a time where, if it was eligible for
copyright protection, it would still be protected by copyright. A good public domain rule of thumb is
anything after 1922 is still protected by copyright, and the technical
realities of 1920s-era recording means any high quality recording is much newer
However, there is another factor to consider. As I am admittedly superaggressive
about mentioning, works by the US government are not eligible for copyright
protection. They are in the public
domain from the moment they are created.
Since the US Navy Band is part of the US government…. Copyright two:
check. Former Public Knowledge intern Ethan Jamescovered
this territory about this time last year.
So What Happened?
If the song (composition and performance) are in the public
domain, how could someone have asserted a copyright claim against it? Enter
YouTube’s Content ID system. Content ID
is a private alternative to the DMCA that YouTube has set up on its site. Large content owners upload music or movies
that they own. Whenever anyone else
uploads a video to YouTube, YouTube checks the video against this blacklist to
see if there is a match. If there is a
match, the rightsholder gets to decide what happens. Sometimes the video is
taken down, sometimes the rightsholder runs ads against it, and sometimes the
rightsholder prevents the uploader from running ads against the video.
There are a number of legitimate criticisms against Content
ID, many of which focus on the fact that it lacks many of the minimal
protections for misidentification built into the DMCA. To continue a theme, former Public Knowledge
Artist in Residency Elisa Kreisinger’s
Use® explores some of these shortcomings (Public Knowledge is on this
In this case, the copyright flag was tied to Rumblefish, a
service that (among other things, I assume) patrols YouTube for copyright
infringement on behalf of its clients.
While the Navy Band’s performance was in the public domain, it sounds like it was close
enough to the (copyright protected) performance by another band represented by
Rumblefish to be flagged.
As an aside, even if this other band’s performance was
identical to the Navy Band’s, it could still be protected by copyright. Copyright allows for independent
creation. As long as the unnamed other
band wasn’t actually copying the Navy Band – but rather getting to the same
place for the same reasons – the Navy Band’s performance could be in the public
domain while the other band’s performance could be protected by copyright. That’s why you have to pay for the London
Philharmonic’s recording of fancy classical music but can also get a copy for
free from Musopen (Public Knowledge
were on the podcast in 2012).
So YouTube’s Content ID confused the Navy Band’s America the
Beautiful with Rumblefish’s client’s America the Beautiful and handed
Rumblefish a veto over Adafruit’s video. Because Adafruit is well known, they were able to raise a stink and have Rumblefish back down. This is a win(ish) for Adafruit, but being big enough to raise a stink really shouldn’t be a prerequisite for overcoming bad copyright claims.
What Could be Done
Better/Where Should I Direct My Rage?
Content ID is imperfect.
That’s inevitable, but knowing that Google and YouTube could work harder
to build safeguards into the system. And
knowing that it represents performances that are dangerously similar to public
domain ones, Rumblefish could work harder to tread lightly.
In this case, one obvious change would be to how Rumblefish
responded to the match. From what I can
tell, Rumblefish had a match to this performance set to “prevent monetization
by the uploader.” While that is better
than “take down forever,” it still had the effect of dictating terms for a
video it had no rights over. For
performances of songs like America the Beautiful – songs that are in the public
domain, that are likely to have multiple similar versions, and that are likely
to have public domain and/or freely licensed versions floating around - Rumblefish could set a match to “flag for me but do nothing until an actual
human being reviews the match.” That is
doubly true because it is much more likely that the free, public domain version
of the song is incorporated into an uploaded video than the one represented by Rumblefish.
Is that even an option in Content ID? I don’t know.
If it is, Rumblefish should take it. If it isn’t YouTube should
implement it. The cost to free
expression of not having that seems a bit too high not to.
The other thing that YouTube could do would be to build some
sort of penalty for false positives into the system. An occasional incorrect match is one thing,
but if a song that is included in Content ID regularly triggers
misidentification perhaps it should not be allowed to claim control over
videos. This kind of common sense
restriction could go a long way towards avoiding these types of situations.