In this article, we discuss the scope and limitations of copyright protection in data. Before diving into the substantive legal issues, it is worth prefacing what data is not. Although often referred to as the ‘new oil’, data is not a commodity that can be ‘sold’ or traded, let alone ‘owned’. Contrary to popular beliefs, the law does not recognise nor confer proprietary rights in data in its own right.
Raw data is also generally not intrinsically valuable. When harnessed correctly, data can create value and inform decision making, including optimising business operations and service delivery. The vast quantities of data have enabled medical breakthroughs and supported the development of new products such as artificial intelligence. Often, the compilation and application of data requires a degree of sophisticated analysis and contextual understanding. It is in this context that data intersects with copyright law, conferring protection to the author’s original contribution.
The basics of copyright
In Australia, copyright is a statutory right established under the Copyright Act 1968 (Cth) (Copyright Act). Unlike trade marks, patents or designs, copyright is not a registrable intellectual property right in Australia. Copyright is a right that exists ipso facto, that is, it subsists in original works recognised by the statute. For a work to attract copyright protection, it has to be an ‘original’ literary, dramatic, musical or artistic work. Unlike the EU where a database can be protected through database rights, which apply regardless of the originality of the database, a database in Australia can be considered a ‘literary work’ but is still subject to the requirement of ‘originality’.
Copyright cannot subsist in a vacuum. This means that the work has to take some material form (ie reduced into an expression). The implication here is that copyright does not protect the idea or information; rather, it protects the expression of that idea or information.1 As such, raw data that exists in its own right without any human input is generally not capable of attracting copyright protection. For example, data about the weather temperature or altitude of a flight path is not information capable of being protected or copyrighted.
Common law jurisprudence has established that the creation of ‘original’ works requires ‘independent intellectual effort’.2 ‘Originality’ in this context does not refer to the novelty of the work, rather, it refers to the ‘origin’ of the work. In other words, a work must ‘originate’ from an author rather than being a mere copy of another work. A work produced by way of an automated process is generally not sufficient to satisfy the requisite element of the ‘originality’. In the case of Telstra v Phone Directories, the trial judge (whose judgment was affirmed in the Full Court of the Federal Court) ruled that copyright does not subsist in phone directories that were compiled with an automated computer process. Gordon J stated:
None of the Works were original. None of the people said to be authors of the Works exercised ‘independent intellectual effort’ or ‘sufficient effort of a literary nature’ in creating the Works. Further, if necessary, the creation of the Works did not involve some ‘creative spark’ or the exercise of the requisite ‘skill and judgment’.3
To work out whether copyright subsists in a work, her Honour articulated the following approach:
…Start with the work. Find its authors. They must have done something, howsoever defined, that can be considered original.4
Accordingly, for copyright to subsist in datasets, it is crucial to ascertain the originality and authorship of the work. Some practical steps that organisations can do to increase the likelihood of a work being accorded copyright protection include documenting the process of the creation of the work, maintaining appropriate records in relation to the work, reducing reliance on automation to create the work, and ensuring sufficient human involvement and contribution in the process as much as possible.
What copyright protects
Copyright, sometimes also known as economic rights, grants the owner of a work a bundle of exclusive rights to do acts comprised in the copyright. In the case of literary work, this includes the right to reproduce the work in a material form, publish the work, communicate the work to the public, and make adaptation of the work.
A person who does an act comprised in the copyright without the authority of the copyright owner infringes the copyright of the owner. An infringing act includes the substantial reproduction of a copyright work, for example, the photocopying of a large portion of a textbook. In determining whether a ‘substantial’ part of a work has been reproduced, the courts tend to also have regard to the qualitative element of the originality of what is copied, and not merely the quantitative aspect of it.
Given copyright protects the expression of the idea or information and not the information or underlying concept itself, it is not always sufficient to rely on copyright protection alone. For example, while it may be a copyright infringement for a third party to reproduce an important dataset without the authority of the copyright owner, applying the knowledge contained in that dataset may not necessarily be a copyright infringement in and of itself, unless the act falls within the exclusive rights comprised in the copyright.
Copyright is not a substitute for confidentiality or non-disclosure agreements. It is simply one of the many mechanisms to protect intellectual property rights.
The case for contracts
Given the limitations of copyright protection, there is often still the case for entering into a data sharing or non-disclosure agreement. This allows parties to agree to the ‘ownership’ of the data, and each party’s rights in respect of the data. For data sharing arrangements, it is also important to consider imposing guardrails around the licence to use and disclose data, and post-termination issues.
Entering into a contractual arrangement also creates a binding obligation between the parties which permits a party to pursue a breach of contract and recover losses or damages where data is used for other purposes. The other benefit of pursuing a breach of contract is that this cause of action tends to be more straightforward compared to a case of copyright infringement, which requires the party asserting copyright infringement to first demonstrate that copyright subsists in the work.
Takeaways
The potential for data creation is virtually unlimited, making it more abundant and renewable than oil. It is therefore crucial for organisations to take steps to ensure that any datasets they create are ‘original’ and involve sufficient human intellectual effort. Organisations should also be aware of the limitations of copyright protection, which should not replace the need for confidentiality arrangements.
How can HWL Ebsworth help?
HWL Ebsworth’s Intellectual Property team has extensive experience in advising businesses on intellectual property issues. If you have any queries about intellectual property law, please do not hesitate to contact us for further information on how we can assist you.
Important Disclaimer: The material contained in this publication is of general nature only and is based on the law as of the date of publication. It is not, nor is intended to be legal advice. If you wish to take any action based on the content of this publication we recommend that you seek professional advice.
This article was written by Luke Dale, Partner, and Paul Sigar, Solicitor.
1 Network Ten Pty Ltd v TCN Channel Nine Pty Ltd (2004) 205 ALR 1 [51].
2 IceTV Pty Ltd v Nine Network Australia Pty Ltd (2009) 254 ALR 386 [33].
3 Telstra Corp Ltd v Phone Directories Co Pty Ltd (2010) 264 ALR 617 [340].
4 Ibid [344].
