A bill before the Australian Parliament will extend immunities from copyright infringement currently enjoyed by internet service providers (ISPs) to cultural institutions, educational institutions, and organisations assisting people with disabilities.
These immunities are known as “safe harbours”. They can be a useful policy tool. They help ensure that exposure to copyright liability doesn’t inhibit socially or economically useful activity such as developing internet infrastructure.
We argue that copyright issues for not-for-profit organisations differ from those affecting ISPs and their subscribers. Rather than expanding existing safe harbours, policy makers should tailor the concept to the needs of the cultural sector.
Safe harbours have been justified as a quid pro quo to encourage investment in Internet infrastructure. ISPs were concerned that they could themselves be liable for their customers’ copyright infringements. Investment in internet technologies might have been inhibited by the risk of potentially enormous damages claims.
Currently, Australian safe harbours apply only to telecommunications providers such as Telstra and Optus. They shelter ISPs against monetary awards for copyright infringement for actions such as transmission and caching, and storage of infringing material uploaded by subscribers. The immunities don’t apply when the ISP is itself responsible for storing infringing material.
Immunity can also be lost when an ISP fails to meet key conditions, such as removing infringing material once the ISP knows about it.
Fresh thinking on safe habours
At first blush the bill makes sense. The institutions mentioned in the Copyright Amendment (Services Providers) Bill 2017 (Cth) do important work. A not-for-profit running a website that allows the uploading of material suitable for people suffering from visual disabilities, for instance, deserves immunity from copyright damages claims if some of that material turns out to be infringing.
However, the Australian bill should also prompt us to revisit the safe harbour concept itself. A safe harbour reflects a policy that certain activities are sufficiently important that they should not be inhibited by the risk of copyright proceedings. Rather than extending existing safe harbours to other institutions, we should look at how such institutions are currently inhibited by exposure to copyright claims, and how the law might be adapted.
There are many ways we can craft copyright immunities for the not-for-profit GLAM sector: public galleries, libraries, archives and museums. GLAM institutions currently enjoy some exemptions from copyright infringement, and there have been calls for a broader fair use or fair dealing defence.
Specific shelter for GLAM sector
Another possibility would be a safe harbour that is tailor-made for the GLAM sector.
The reforms in the Australian bill would apply only where the infringing activity is undertaken by patrons. But unlike ISPs, GLAM institutions need the greatest protection for the things they do themselves. We value museums, for example, for the expert work of their own curators and collection managers, not because they sometimes allow other people to upload material to their websites.
GLAM organisations use digital technologies in innovative ways. They put their collections online, link to the online collections of other institutions, and create vast repositories of metadata about collection items. Some exhibitions are digitally curated, with links to related objects and information, often located on remote sites. While some GLAM organisations allow third parties to post material to their websites, that’s not their core work.
Anyone working in the GLAM sector will tell you how copyright concerns limit their work. Institutions also fear the reputational damage of allegations that they disregard copyright. But so long as GLAM organisations act responsibly, these valuable contributions to cultural life should not be unduly inhibited by risks of copyright liability.
Striking a balance
Most GLAM sector organisations are responsible about copyright. If GLAM institutions use some material without copyright permissions, this is typically of limited commercial significance – quite unlike the commercially valuable material that is transmitted by ISP subscribers every second.
A fit-for-purpose GLAM safe harbour could encourage responsible behaviour while ensuring that the GLAM sector’s work is less impeded by risks of copyright claims. Unlike the ISP scheme, a GLAM safe harbour would not be limited to patrons’ activities. It might be conditioned on good faith efforts to raise copyright awareness among staff, including regular training. In the digital context, it might have a take down requirement.
In summary, the copyright issues for publicly funded not-for-profits doing valuable social work are not the same as those affecting ISPs and their subscribers. Lumping them together in the same safe harbour avoids more nuanced thinking about their relative social value and the different risks each poses for those who rely on copyright protections for their livelihood.
Australian creators struggle to understand copyright law and how to manage it for their own projects. Indeed, a new study has found copyright law can act as a deterrent to creation, rather than an incentive for it.
Interviews with 29 Australian creators, including documentary filmmakers, writers, musicians and visual artists, sought to understand how they reuse existing content to create. It considered issues such as whether permission (“licences”) had been sought to reuse copyrighted content; the amount of time and cost involved in obtaining such permissions; and a creator’s recourse if permission was either denied or too expensive to obtain.
For the majority interviewed, seeking permission to reuse copyrighted content – for example, as snippets of music or video in films, or long quotes in written works – was a source of great frustration and confusion. The process was variously described as “incredibly stressful”, “terrifying” and “a total legal nightmare”.
Problems mostly centred on time delays and financial expenses. Creators found that the paperwork required to request permission was often long, complex and not standard across publishers and other rights-holder bodies. Many waited months for a response to a request; some never received one at all. Many reported feeling ignored and disrespected.
One interviewee, a composer, waited over a year for permission to set poetry to music. The music was due to be performed in a theatre production. The original poet was deceased but his publisher controlled the copyright.
After waiting months and not receiving a response, the composer was forced to painstakingly replace the words to the song with new ones that fit the same rhyme scheme, stresses, cadences and meaning as the original poem. This was a long and difficult process. Roughly a year after the play was staged, permission to use the poem came through from the publishers. By then it was too late.
Licence fees were also an issue for the creators interviewed. Licence fees can be expensive, even for very small samples. Many creators thought that copyright fees demanded for reusing small samples were unfair and stifling.
A filmmaker making a documentary about a small choir in rural Australia could not afford the licence fees to release the film to the public. To show snippets of songs sung by the choir, totalling less than two minutes of copyrighted music in a 20-minute film, with each snippet only seconds in length, the licence fees came to over $10,000. The project was ultimately abandoned because the filmmaker could not raise the funds to cover the licensing fees.
Avoiding and abandoning projects were common reactions to the restraints imposed by copyright law, although a very small number of creators proceeded anyway, hoping to “fly under the radar”.
Some changed projects to try to circumvent copyright restrictions. For example, filmmakers might degrade the sound on their films for scenes where background music might be playing, such as those filmed in a pub or restaurant.
Ideas were filtered out early at the brainstorming stage because they were “too risky” or licensing would be “too expensive”. Some people avoided entire areas of creativity, such as appropriation art, music sampling or documentaries about music or musicians, because it was all just “too hard”.
Court decisions such as the 2010 “Kookaburra” case have further aggravated the problems. In this case, despite significant elements of original creativity, the Australian band Men at Work were found to have infringed copyright of a 1934 folk song, Kookaburra Sits in the Old Gum Tree.
This case is a classic example of the gap that exists between law and creative norms. The law’s concern, in that case and others, is with what has been taken from an existing work. Creators, on the other hand, most commonly focus on the elements they have added to the work.
The study also highlights creators’ confusion about the scope and application of Australian copyright law. Creators were especially confused about legal exceptions to copyright infringement. In Australia, these are called “fair dealing” exceptions and they are narrow – they apply only to specified purposes (such as for research and study; parody and satire; reporting the news; and criticism and review).
Creators expressed concern about what, exactly, fell within “parody and satire” or “criticism or review”. What do those terms mean when applied to art? Once participant remarked: “Everybody is out there flying a bit blind about this.”
Other countries, including the United States, South Korea and Sri Lanka, have broader exceptions to copyright infringement, which permit reuse for things such as remix or appropriation art, provided that the use is “fair”. These exceptions are generally called “fair use”. Importantly, these exceptions do not require the use to fall within a predetermined category, like reporting the news. Each use is assessed on its own merits.
Courts apply some basic standards in determining what amounts to “fair use”, which include examining the purpose for which an original work has been used; the extent to which it has been transformed; and the extent to which a new work impacts on the market of the original work.
In recent years, the Australian Law Reform Commission and Productivity Commission’s recommendations that Australia adopt a US-style fair use exception attracted significant criticism from much of Australia’s creative sector. Many considered that such an exception would be too broad and too uncertain. However, the study suggests this criticism may be largely unfounded.
The creators interviewed used their own strong sense of morality and fairness to guide what reuse they considered to be acceptable. These principles and norms align quite closely with the factors that courts use in assessing fair use, including how much new creativity has been added to the existing work and whether the new work commercially impacts the existing work in an unfair way.
This new study suggests that more flexibility in the law might actually help to spur the creation of new Australian work.
Safe harbours protect internet hosts and platform providers from monetary liability for copyright-infringing content posted or shared by their users. For example, if you post the latest Thor movie to YouTube, YouTube won’t be responsible for copyright infringement if it takes down that video. In Australia, we only extend this protection to internet services providers, not general purpose websites.
This matters because technology firms rely on limits to liability to manage their risks. Companies like Facebook or YouTube, which host millions of pieces of user content, would face serious difficulty starting in Australia because our laws on copyright infringement are so strict.
The new legislation is a step in the right direction, but it doesn’t go far enough to create an environment that fosters Australian innovation.
Excluding platforms from safe harbours doesn’t make much difference to tech giants like YouTube and Facebook, since they already operate within the United States safe harbours. But it does discourage Australian tech start-ups from the chance to experiment in a reduced-risk environment.
It is not just the US with broader copyright safe harbours than Australia – jurisdictions around the world extend safe harbours to internet intermediaries beyond ISPs.
The European Union, for example, provides that member states must ensure that any hosting provider will not be liable for unlawful content posted by users, provided it acts quickly to remove the content upon notice.
The copyright safe harbours came about as a result of the US Digital Millennium Copyright Act (DMCA) in 1998. The DMCA represented an important bargain struck between the established content industry, such as big film and TV studios, and the burgeoning tech industry.
The content industry got a “notice-and-takedown” regime that required online service providers to remove material that infringes copyright. In exchange, the tech industry got copyright safe harbours.
Under this system, the service provider must quickly and efficiently remove infringing content if they are informed about it by the copyright owner. This notice-and-takedown scheme has become fundamentally important to the way the internet works today.
But in enacting the copyright safe harbours, parliament made a drafting error. Instead of extending protection to “service providers”, as the US law does, we gave protection to “carriage service providers” as defined in the Telecommunications Act.
Essentially, Australia only gave protection to internet service providers like Telstra, Optus and TPG, and not to platform providers like Whirlpool, RedBubble, YouTube or Facebook. For more than a decade, this has been a critical difference between US and Australian copyright law.
The new bill appears to close the glaring gap between US and Australian law by replacing the term “carriage service provider” with, simply, “service provider”.
But the bill defines “service provider” to be either a carriage service provider; an organisation assisting persons with a disability; or a body administering a library, archives, cultural institution or educational institution.
It does not extend the safe harbour to those who actually need it the most – Australia’s internet hosts and platform providers.
This is a seriously missed opportunity for Australian innovators. There is a real risk for businesses, both large and small, who want to provide online spaces for people to communicate.
Our copyright laws potentially make hosts liable for much of the copyright infringing content that users may upload or share. But it can be prohibitively expensive and time-consuming to pre-screen all content before it is uploaded.
This is one of the reasons why many large social media platforms don’t base their operations in countries like Australia, and why Australian businesses are at a major competitive disadvantage compared to those in other countries.
Why not extend the safe harbour to Australian innovators?
There were early indications that the Australian government intended to extend the safe harbours to all online service providers, but these amendments were shelved.
Entertainment industry groups have been lobbying hard in recent years for measures that go beyond the notice-and-takedown scheme that the safe harbours provide. They want what they call notice-and-staydown: proactive filtering of unlicensed copyright content by service providers.
At the same time, copyright owners want higher payments. They use the term “value gap” to describe what they see as the difference between sites like Spotify that pay hefty licence fees to make content available to users and sites like YouTube that do not.
Content owners are no longer happy with the bargain they struck in the DMCA – they allege that sites like YouTube are gaming the system of the safe harbours.
There is a false equivalency at work here. Spotify is not a site for user-generated content and does not purport to be; sites like YouTube have everyday users at their core. If we believe that creative discourse, engagement and play matters then there is a cogent reason why sites that facilitate user-generated content might need some legal latitude.
However, this debate misses a more fundamental point. Limited safe harbour provisions hurt Australian creators and innovators. They increase the risk to innovators developing new technology products and platforms.
And, importantly, Australian creators miss the opportunity to exercise greater control over their creations through notice-and-takedown mechanisms that are easy to use and far cheaper than copyright lawsuits.
Copyright law sometimes allows you to use someone else’s work – as long as it’s fair. In Australia this is called “fair dealing”, and it’s different to the law in the US, which is called “fair use”.
These exceptions are safety valves in copyright law – they allow lots of beneficial uses that society has agreed copyright owners should not be able to charge for, or worse, prevent.
There’s a serious ongoing debate about whether Australia should update its copyright laws and introduce fair use. The current law is not easy to understand – our research shows that Australian creators are often confused about their rights – and many think we already have fair use.
Fair dealing: What can you do in Australia?
The key difference between “fair use” and “fair dealing” is that Australia’s “fair dealing” laws set out defined categories of acceptable uses. As we will see, “fair use” in the US is much more flexible.
Australian copyright law sets out five situations where use of copyrighted material without permission may be allowed:
research or study
criticism or review
parody or satire
reporting the news
provision of legal advice.
We’ll explain the first four, as they’re most useful to the average Australian.
Research or study
You do not need permission to copy a reasonable portion of copyrighted material if you are studying it or using it for research. You do not have to be enrolled in school or a university course to rely on the research or study exception.
you can make a copy of a chapter of a book to study it
you can print or take screenshots of content you find on the web for your research
you can include quotes or extracts of other work when you publish your research.
The main thing to watch out for is how much you copy. It’s fair to photocopy a book chapter but not the whole book.
Criticism or review
It is lawful to use a work without permission in order to critique or review it.
Criticism or review involves making an analysis or judgement of the material or its underlying ideas. It may be expressed in an entertaining way, or with strong opinion, and does not need to be a balanced expression to be fair.
For example, a film critic does not need permission to play a short clip from a film they are reviewing. They may also use film clips from other movies to compare or contrast.
It’s also legal to quote an excerpt of a book or song lyrics, or to reference a photograph in another publication as part of a review or critique of the work.
You need to be really critiquing your source material. So, for example, a review video that is really just the highlights of a film or show probably won’t be fair.
This is something that tripped up Channel 10 in its clip show, The Panel. When the panellists discussed and critiqued the clips they showed, it was generally fair dealing. But when they just showed clips that were funny, a court found them liable for copyright infringement.
Reporting the news
You don’t need permission to use existing copyrighted material while reporting on current or historic events. The law is designed to ensure that people can’t use copyright to stifle the flow of information on matters of public interest.
The key issue to check here is whether a work has been used in a way that is necessary to report the news. If the material is just used incidentally, to illustrate a story or provide entertainment, it won’t count as fair dealing.
Parody or satire
It is legal to use another person’s copyrighted material without their permission to make fun of them, or to make fun of another person or issue.
Making something funny is not sufficient to rely on this exception. The use must be part of some commentary (express or implied) on the material or some broader aspect of society.
When is a use ‘fair’?
Fair dealing only applies when the use is “fair”.
When assessing fairness in Australia, there are a number of relevant considerations, including:
how important copying is to your work (“nature and purpose of the use”)
the type of work being copied (less original works may not be protected as strongly as more creative works)
whether it is easily possible to get a licence within a reasonable time at an ordinary commercial price
the effect of your copying on the potential market for the original
the amount taken from the original work
whether attribution has been given to the original author.
Generally, a use will be fair if you are copying for a valid reason, you don’t copy more than you need, you give attribution where possible, and your work is not directly competing in the market against the original.
Things to remember:
Is copying necessary? Copying has to be necessary for one of the purposes above. This means that it might be fair to copy part of a song to review it, but it won’t be fair if you’re just using the song as background music.
Copy no more than you need. Sometimes you need to copy the entirety of an existing work – if you’re critiquing a photograph, for example. Usually, though, you should only copy the parts that are necessary. You can’t get away with showing a whole TV episode in order to critique one scene.
It’s usually not fair if you’re competing with the original. This is often the most important factor. When you copy existing material for your own study, to report on the news, or to create a parody, you usually won’t be undercutting the market for the original. But if you’re just repackaging the original material in a way that might substitute for it – a consumer might be satisfied with your work instead of the original – then your use probably won’t be fair.
How is ‘fair use’ different – what can’t you do with fair dealing?
In the United States, the law is more flexible, because it can adapt to allow fair use for purposes that lawmakers hadn’t thought of in advance.
Some of the things that are legal without getting permission in the US but not in Australia include:
Adapting to new technologies: Fair use is flexible enough to adapt to change, but fair dealing is not. For example, in the US, fair use made it legal to use a VCR to record television at home in 1984. In Australia, this wasn’t legal until parliament created a specific exception in 2006 – just about the time VCRs became obsolete.
Artistic use: In Australia, it’s legal to create a parody or a critique, but not to use existing works for purely artistic purposes. For example, Australian law makes it largely unlawful for a collage artist to reuse existing copyright material to create something new.
Uses that document our experiences: Media forms a big part of our lives, and when we share our daily experiences, we will often include copyright material in some way. Without fair use, even capturing a poster on a wall behind you when you take a selfie could infringe copyright.
In a famous example, Stephanie Lenz originally had an adorable 29-second clip of her baby dancing to a Prince song removed from YouTube, due to her use of the song. She was able to get it put back up under US fair use law – but an Australian wouldn’t have that right.
Technical and non-consumptive uses: The internet we love today is built on fair use. When search engines crawl the web, making a copy of every page they can in order to help us find relevant information, they’re relying on fair use.
Under Australian law, even forwarding an email without permission could be an infringement of copyright.
It’s been suggested that introducing fair use here would provoke a “free for all” use of copyrighted work, but that hasn’t happened in the US. In fact, some of the same major studios that oppose fair use in Australia are at pains to point out that they support fair use in the US because it is vital to commercial production that happens there.
The Motion Picture Association of America, for example, says that “Our members rely on the fair use doctrine every day when producing their movies and television shows”.
To put it simply: we don’t think that fair use will harm creators.
The “fair” in fair use means that it’s not about ripping off creators – it mainly allows uses that are not harmful. But we do think that fair use would provide an important benefit for ordinary Australians – both creators and users.
Katherine Gough, a musician and law student at Queensland University of Technology, co-authored this article.