National and International Dimensions of Copyright’s Public Domain (An Australian Case Study)

Graham Greenleaf *

1. The value of Australia’s copyright public domain

1.1 Introduction: What do these innovations have in common?

A set of examples follow. They all involve valuable contributions to Australian innovation in the area of information goods. They all involve copyright works in which various parties have continuing proprietary (copyright) interests. But they also involve the public (or classes of the public) having rights to use those works in ways that involve some of the exclusive rights of the copyright owner. In this article, as discussed in Part 2, I use the expression “public domain” in the expansive sense of encompassing all “public rights” in copyright. “Public rights” are all those aspects of copyright law and practice that are important in determining the ability of the public (or a significant class of the public) to use works without obtaining a licence on terms set (and changeable) by the copyright owner.

The Creative Commons slogan “Some Rights Reserved” sums up rather well the way in which the intellectual goods in the following examples combine proprietary and non-proprietary elements. However, most of the examples of what I will call “the public domain” do not involve the use of Creative Commons licences.

The theme of this article is what these examples have in common, how Australia’s copyright law and the institutions that support innovation have paid insufficient attention to what they have in common, and how – in Australia at least – we need to have a law reform review that will have these common elements (the copyright “public domain”) as its focus.

After introducing these examples, I give some further consideration to the terms “public domain” and “public rights” (Part 2), and consider how the public domain in any particular country comprises both “global” and “national” elements (Part 3), with examples of what makes Australia’s public domain distinctive (Part 4). In the second half of the article (Parts 5-13) I consider in detail eight main aspects of how a holistic approach to Australia’s copyright public domain suggests areas in need of reform, and some of the interconnections between them. From this I conclude (Part 14) that there needs to be a law reform review of Australia’s copyright law with the public rights / the public domain as its focus.

1.2 Examples from Down Under

Australians continue to make very substantial contributions to the development of open source software and thus to the Internet’s global infrastructure. Contributions to the Linux kernel have included the port of Linux to the PowerPC architecture (largely done in Canberra); the work to put Linux on Cell processors; and contributions to the Sparc processor work. Australians hold senior positions in overall Linux development, and in subsystem maintenance. Australian work on Linux networking is the basis for many companies building firewall appliances, smart routers etc. The Samba re-implementation of the SMB/CIFS networking protocol,1 initially developed at the Australian National University, has been the basis on which various companies have built their businesses. The file transfer utility rsync2 has similar antecedents. Other major contributions include the pppd dameon used by a significant proportion of ADSL home routers; the radiud authentication server used by many ISPs; the SSL library which is the cryptography engine used by much e-business (developed in Queensland); contributors to the Firefox browser; gcc use in the embedded computing market; and much development of the Gnome and KDE Linux desktop environments which are becoming increasingly important.3

The AEShareNet Licensing System,4 operated by TVET Australia, licences about 3,000 learning objects for free educational use,5 and in some cases with rights to modify, primarily for use in the technical and further education (TAFE) sector. In addition, about 600 pages on the web use its “Free for Education” (FfE) licence.6 The AEShareNet licence suite was one of the world’s earliest developments of open content licensing. AEShareNet resources are searchable along with other Australian educational resources from all sectors via Education Network Australia (edna)7 but it is not possible to limit searches there to items that are available for free educational use or modification.

PANDORA,8 described as “Australia's Web Archive” by the National Library, has developed since 1996 an expanding collection of selected Australian online publications, such as electronic journals, government publications, and web sites of research or cultural significance. Built in collaboration with nine other Australian libraries and cultural collecting organisations, PANDORA now has over 50M files (over 2 TB) comprising over 36,000 “archived instances.” It is growing at a rate of nearly 2% per month. PANDORA is archiving the history of Australia’s “online search commons.”

Australia has played a significant role in the development of e-learning platforms. LAMS (Learning Activity Management System)9 is open source software developed at Macquarie University which provides a visual tool for designing, managing and delivering online collaborative learning activities. It has been adopted in 2007 by the Northern Territory Department of Employment, Education and Training for roll-out in all of its schools.10 Moodle11 is an open source software e-learning platform, with particular strength in wiki development, and with a world-wide network of developers coordinated by the Moodle company based in Perth, Western Australia. It has a user base of 42,080 registered sites12 with 16,927,590 users in 1,713,438 courses.13

Usage of Creative Commons licences in relation to Australian content had resulted in at least 2,000 websites containing over 100,000 pages of Creative Commons licensed content as of early 2007, though this may be something of an under-estimate.14 This includes both the use of the Australian Creative Commons licences, and the use of the “generic” CC licences over Australian materials, as shown in the graph at right.15 The most popular licence attributes are “Non-commercial” and “Share-alike”. A few of the best known sites using Creative Commons licences are noted in the examples below.

Despite Australia’s Crown copyright in cases and legislation, Australian law has been the site of world-leading initiatives in free access to law. The NSW government gazetted a general licence to the public allowing reproduction of legislation and case law as far back as the 1990’s. The Australian Council of Chief Justices’ “Court designated citations” standard, adopted in 1998, has enabled cases to be authoritatively cited from the moment they are handed down. This standard has been adopted in the United Kingdom (UK), southern Africa, the Pacific Islands and elsewhere across the common law world. Since 1995 the Australasian Legal Information Institute (AustLII), a community service and research infrastructure initiative of two university law schools, has created the world’s largest free access online law resources16, including such features as free Point-in-Time legislation, and the world’s only comprehensive national treaties collection. Over 200 public institutions, including courts, tribunals, government agencies and universities, have collaborated with AustLII to produce these free access repositories. AustLII’s open source Sino search engine is used by the majority of free access “legal information institutes” around the world.

Screenrights, the collecting society for educational use of audio-visual materials, takes an innovative approach to educator’s public rights to use a-v materials. It regards this as a business opportunity for its members: “[w]ith Screenrights now collecting more than $23 million a year from educational copying, filmmakers are becoming increasingly aware of the benefits of marketing their work effectively to this sector.”17 As part of this business model, it contributes back a large quantity of free-access educational resources such as study guides, about programs which are being screened on TV , through its Enhance TV service: “[e]ducational programs to be broadcast on television are highlighted on the site, and in an email newsletter circulated to over 20,000 Australian educators every week. This gives teachers the advance notice they need to copy, and provides exposure for filmmakers.”18

Free access repositories of Australian academic research are becoming more common. Institution-based repositories are being established at various Australian Universities, including those repositories from members of the ARROW19 consortium (Australian Research Repositories Online to the World), which between them currently provide about 12,000 academic papers for free access (but usually with no licences providing other rights).20 Swinburne University of Technology has the largest repository (Swinburne Research Bank)21 with 7,003 papers, followed by Monash University ARROW Repository22 (2,516 papers), and the University of the Sunshine Coast “Coast Research Database”23 (1,933 papers). Of the ARROW databases, only the smallest, the University of New South Wales UNSWorks24 (348 papers) requires depositing authors to use a Creative Commons licence (Attribution-Noncommercial-NoDerivatives). There are significant institutional repositories outside ARROW,25 such as Bond University’s e-publications@bond (1,856 papers),26 QUT ePrints,27 University of Wollongong’s Research Online (2,064 papers),28 Flinders Academic Commons29 (1,525 papers). In addition, national discipline-based repositories are starting to develop. AustLII provides free access to over fifty Australasian law journals (6,576 papers),30 and has received ARC funding to expand this into a national Legal Scholarship Library.31

The National Library of Australia’s Picture Australia32 aims to be the definitive pictorial service for and about Australians and Australia, providing one search over collections in 45 major Australian public institutions. Pictures (photos, sketches, cartoons etc) are often only available for private research and study, but some are available for other uses.33 Picture Australia’s “Click and Flick” is an initiative to open Picture Australia to contributions from the Australian public, through uploads to Flickr using Creative Commons licences.34 Picture Australia now includes over 1.1 million images from the collections of forty-five organisations and, now, individuals via Flickr. Its federated combination of public domain images, Crown copyright images made available for free access, and images contributed by the public under voluntary public rights licences (Creative Commons) is indicative of what creative collaboration among public institutions can do.

Other major cultural institutions are starting similar initiatives. The Powerhouse Museum, Sydney is one of the first cultural institutions in the world to release its public domain images on Flickr35 (April 2008), following the example of the United States Library of Congress (launched January 2008). The Powerhouse is releasing its Tyrrell Collection, 7,903 glass plate negatives recording life in Australia around the turn of the twentieth century, of which 300 have initially been included on Flickr.36 A statement accompanies each image: that they have “no known restrictions on publication”. The example here shows Pyrmont Bridge, Sydney, at the turn of the century.37

Sites supporting creativity that use open content licences are increasing. Multimedia site 60Sox38 provides a focal point for emergent creativity in Australia and New Zealand and uses Creative Commons Australia licences. Australian Creative Resources Online (ACRO)39 hosts a repository for audio, video, and still images that can be used freely and legally for creating digital art, education, or other uses. ACRO uses both Creative Commons Australia licences and AEShareNet Free for Education licenses. The Brisbane Media Map is a database of hundreds of creative industry organisations in the Brisbane area, linked to Google maps locations, and with its content, database and software all licensed under Creative Commons Australian licences (CC-BY-NC-ND 2.1).40

Finding Australian works in which copyright has expired (the narrow meaning of “public domain”) is harder than one might expect, because there is often no way to search major catalogues (such as at the National Library) to specify only works that are in the public domain. Nor do overseas collections, such as Google Books, which include Australian public domain materials allow searches by that characteristic. Only a few specific “out of copyright” Australian collections exists. The University of Adelaide Library’s eBooks@Adelaide41 (969 records) includes more than 500 classic works of literature, philosophy, science and history (not necessarily Australian) in the public domain. Project Gutenberg Australia42 is a private initiative which includes numerous Australian works (and other works) in the public domain, and can be searched using Google’s search engine, but it is very difficult to gain any clear idea of its coverage.

Australian scientists are making use of publication in Australian or international free access research repositories, and are sometimes utilising open content licences with re-use rights. A Murdoch University research team published their simple diagnostic test for African Sleeping Sickness43 with the Public Library of Science under an Australian Creative Commons licence (CC BY 2.5).

Politics and public affairs websites make use of Creative Commons licences. YouDecide2007,44 a citizen journalism initiative between SBS, On Line Opinion, the Brisbane Institute, and QUT Creative Industries covered the 2007 Australian federal election, using the CC BY-NC-ND 2.5 AU licence. On Line Opinion45 uses the CC BY-NC-ND 2.0 licence. EngageMedia, a video-sharing website, focuses on social justice and environment issues in Australia, South East Asia and the Pacific.46

Civil society organisations are also making increasing use of open content tools. The Association for Progressive Communications Australia47 has released ten years of documentation on the use of ITC for community development, under a Creative Commons licence (CC BY-NC-ND 2.5 AU) on a publicly-available wiki for Document Freedom Day 2008.48

1.3 An important caveat

These examples of Australian intellectual goods illustrate the major role that public rights of differing types play in Australian innovation and culture. This selection of public resources is only the tip of the iceberg, both of what Australia’s public domain already comprises and (more importantly) of what our public domain could be in future.

This article is about these public rights and how they can be used to stimulate innovation and enhance national culture. But there is an important caveat which needs to be stated at the outset. Both now and in future such works which involve predominance of public rights over private rights will only be part of Australia’s creative landscape, and most probably the minor part: “[a]ll rights reserved” may continue to be appropriate for most creators, most of the time.

In Australia, changes to copyright law over the previous decade have been principally about strengthening copyright (“all rights reserved”), and have involved much less to stimulate Australia’s public domain (“some rights reserved”). The argument of this article is that we should be trying to get the most out of both methods of stimulating innovation. Other counties may have a similar need.

2. “Public rights” and the public domain

I have been using the terms “public rights” and “public domain” almost interchangeably. I will explain shortly more precisely what I mean by “public rights”. “Public domain” is an ambiguous term. In its narrowest use it means those works in which copyright has expired due to the expiry of the copyright term. A slightly broader usage includes works that do not ever attract copyright protection, and those over which the author has renounced all claims of copyright. However, it has a more modern and expansive usage which encompasses all types of “public rights”, including, in addition to the two uses already mentioned, other aspects of copyright law which give rights to the public, such as fair dealing exceptions and uses allowed under compulsory licences (and other examples given in the next section). “Public domain” in this broad sense can be used in relation to copyright49 and in relation to other forms of intellectual property rights. Samuelson provides seven different maps indicating how different scholars see this broader notion of the public domain.50 In this article, I will use “public domain” as a synonym for the broadest usage of “public rights” in relation to copyright. To avoid confusion, I will also avoid using the word “commons”, though much of the discussion could also take place using that term.

2.1 Avoiding a misleading dichotomy

If the term “public domain” is used in its narrow sense, there is a dichotomy of works: some works are in the public domain but only if they are old enough, and all other works are not. The approach taken in this article is very different, and recognises that almost all works contain both public and proprietary components, and fall somewhere on a continuum between the extremes of works which are subject to “private rights only” and those which are subject to “public rights only”. At the public rights extreme it is easy enough to envisage public domain works such as the plays of Shakespeare.51 But for practical purposes all proprietary works are at least subject to some minimal “fair use” exceptions in copyright law, and so are subject to some public rights. The proprietary extreme of the continuum is empty.52

The normal nature of works is to be a composite of public and proprietary rights, with each work situated at some point along a continuum between the two extremes. There is therefore normally a dichotomy between public and private rights, but it is one which exists within each work, rather than between works. The Creative Commons slogan “some rights reserved” recognises this inherent duality in works and builds on it.

2.2 “Effective” public rights – a useful refinement

For the purposes of this paper, I mean by “public rights”53 all those aspects of copyright law and practice that are important in determining the ability of the public (or a significant class of the public) to use works without obtaining a licence on terms set (and changeable) by the copyright owner. In other words, public rights are “The effective extent to which I can use a copyright owner’s works without seeking the owner’s permission.” Including things that determine the effective exercise of otherwise formal rights reduces the precision of the notion of public rights, but does allow us to give a richer and more useful description of a country’s public domain.

The corollary of this definition of “public rights” is that private/proprietary rights are the effective ability of the owners of copyright in works to refuse to allow other people to use those works, except on terms set (and changeable) by them.

3. The global copyright public domain and Australia

There are elements which are common to the public domains of the vast majority of jurisdictions around the world, principally because of two factors: (i) the near-universal adoption of the Berne Convention (1886) (and some effects of the TRIPS Agreement) and (ii) some effects of the Internet are global, particularly those associated with search engines and with viral licences. To some extent it makes sense to talk about a global public domain, but what is relevant here is the effect of these factors on the shape of Australia’s national public domain.

3.1 International agreements and their limitations on national public domain

The main effect of international agreements has been to restrict what can be included in a national public domain according to international law. The Berne Convention (1886)54 and its subsequent amendments55 can be seen as the main factor responsible for determining the size of national public domains, and many of their features.56 Five specific negative elements constrain public domains.

First, the most significant determinant is that in accordance with Berne, registration of works is not required for copyright (Article 18 and elsewhere). This creates a shrunken public domain, as it reverses the default condition of a work from “public” to “proprietary”. If registration is required, then it can be expected that most works will not be registered, and the public domain will be correspondingly large. When from 1978-89 the United States (USA) abandoned a compulsory registration system for copyright57 and publication and notice requirements and belatedly joined the Berne Convention (1989), this may have been the largest contraction (since Berne itself) of the scope of the world’s public domain. The absence of any requirement for registration is a major contributor to problems such as “orphan works”, where it is impossible to locate a copyright owner of a book that is out of print, but the work is still protected by copyright and therefore cannot be reproduced by others. When the United States had a registration system, only about 10% of all works registered were re-registered at the end of the twenty-eight-year term (even though the cost of renewal was small),58 so the other 90% of those works would then have entered the public domain in the pre-Berne United States environment.

The second restrictive element of Berne is the requirement of a minimum copyright term of life of author plus fifty years (Article 7(1)).59 Without this requirement, the minimum term of protection of some types of works, such as computer programs, may have been less, and they would have become part of the public domain (in the narrow sense) earlier.

Third, moral rights (including integrity) may be perpetual according to the interpretation of some state parties such as France.60 This limits the potential re-use of such works.

Fourth, the provisos to Article 9(2)61 concerning national discretion to allowable exceptions in “certain special cases” (often called the “three step test”) is seen by some as limiting the extent to which countries can expand the scope of fair use and compulsory licences, although this interpretation is highly contested (as discussed in 6.1 following). The TRIPS Agreement62 is mainly relevant to determining the nature of the public domain because it adds another version of the “three step test”.

Fifth, it is a minimum rights treaty (Article 20) and it is therefore possible for a consensus of national developments to further reduce most public rights.

There are also aspects of the Berne Convention that support the existence of public rights, or are at least neutral. These include that there shall not be protection of “news of the day” or “miscellaneous facts” (Article 2(8)). Berne leaves open for national decision (and is therefore neutral on) the vital areas of (a) compulsory licensing; (b) fair dealing and (c) protection of legal materials.63 Perhaps the most important and positive aspect of Berne is something that it does not mention, and therefore leaves outside the scope of copyright, namely that the rights to control who can read, listen to or view a work are not in themselves part of the exclusive rights of copyright owners (although these “user rights” have been made less secure by the recent laws concerning technological protection measures).

In relation to Australia, the Australia-United States Free Trade Agreement 2004 (the AUSFTA) is a bilateral agreement which has required Australia to restrict its public domain. The AUSFTA, formally passed in 2004, included requirements for the extension of the duration of protection beyond the Berne-minimum of life of the author plus fifty years, to a seventy-year posthumous period of protection, and stronger provisions for acts of circumvention involving technological protection measures attached to a digital work by the owner of the copyright.64

It is unlikely that these international agreements will be amended in the short term; indeed, the Berne Convention has not been amended since 1979, long before the rise of digital technologies. It remains one of the most significant treaties in intellectual property regulation, particularly when viewed from a public domain perspective. The obligations in these treaties that restrict the scope of Australia’s public domain are constraints on any Australian law reform. Some of these provisions may be undesirable but they are likely to be permanent. They are the settled context of our public domain. The challenge for those who wish to encourage innovation and the public interest through a broader public domain is to identify practical changes to the law which are consistent with these constraints, or reforms to policy and practices which do not require changes to the law.

3.2 The expanding informal global commons

In contrast to international agreements, global practices related to the Internet have effectively expanded public rights globally, and have therefore expanded Australia’s public domain.

The Internet’s world-wide-web from the early 1990s created a global commons for browsing and private use (including reproduction) of works that authors made accessible on the Internet. From 1996 search engines have created a global de facto commons for the searching of such works. Creation of the searchable commons has required the acquiescence of copyright owners in practices by search engine providers (particularly creation of concordances/indexes and retained caches) which may breach copyright laws on a massive scale, though this varies between countries. This is probably the largest expansion of the effective public domain to occur, at least since the development of public libraries turned the right to read works into an effective public domain. It has been described as an example of creation of a commons by “friendly appropriation and acquiescence.”65 An unresolved question at this stage is whether Google Books and other book search facilities will succeed in creating another extension, a global searchable commons for literature which copyright owners have not made freely accessible via the Internet. This now looks more likely, but the proposed Google Book Settlement66 is not yet finalised.

Viral licences67 are voluntary licences of works offered by the author of the work which allow the software or document to which they apply to be modified or combined with other software or texts, but only on the basis that the resulting software or text is available to the public under the same licence conditions. As a result, where they are adopted in preference to non-viral licences, the quantity of software or texts available to the public under the licence expands. The most effective viral licences create an intellectually very significant and rapidly expanding global public domain in certain types of information. The most obvious and important example is open source software created by the viral GNU General Public Licence (GPL) and some other Free and Open Source (FOSS) licences, There are many millions of instances of such licences being used identifiable globally, as detailed in Part 10. The most important examples in relation to textual works are Wikipedia and other collaborative reference works created under the viral GNU Free Documentation License or similar viral licences.

There has also been widespread adoption across the world of other open content licences which are not “viral”, such as those Creative Commons licences which do not include the “share-alike” attribute. Such licences expand the public domain by allowing either any licencees, or defined classes of licencees, to use the content without payment, according to terms of the (non-viral) licence. Some of these licences have national origins, are tailored to national laws, and are mainly used within a particular country (for example, the TVET/AESN licences in Australia).68 However, the greatest proliferation of content licensed under (non-viral) open content licences is of a “global” rather than national nature. The Creative Commons “movement” and its suite of licences originated in the United States and were tailored to USA law, but have been “ported” to comply with the legal environments of different countries. These suites of “CC” licences have a very high degree of similarity to the “generic” Creative Commons licences, and to those of other countries.

From the perspective of encouraging innovation through public rights, the issues raised by these developments are (i) “are there changes to Australian law needed to ensure that the de facto commons created by search engines is not at risk in Australia?”; (ii) “what changes if any to Australian law are needed to ensure that voluntary licences (viral and non-viral) creating public rights are effective, irrespective of which licences are used to create these rights?”

4. Australia’s national copyright public domain

4.1 A very brief history69

To understand how Australia’s public domain can contribute to innovation, it is necessary to appreciate all of the different types of public rights that comprise this public domain. We also need to note those elements found in other countries’ public domains but which are lacking in Australia. Given that some of the shape of Australia’s public domain has been determined by its own history, and its interaction with international influences, we need to start with an overview of that history. We then move to an analysis of the components of Australia’s public domain, in the expanded sense previously discussed.

In some form or another, the public domain has always existed in Australia. Its shape and boundaries have changed over time, but it has always been a feature of Australian copyright law. While Atkinson has noted that the rhetoric of a balancing of rights or strong public interests considerations have never been at the core of our national copyright legislation,70 they have undeniably played a role in the development of these laws. Such concerns have been reflected in the decision-making of both the legislature and judiciary.

Since the construction of the first colonial copyright legislation, in Victoria in 1869,71 the public domain has been recognised as an issue. It is clear, from this earliest statute, that governments recognised copyright was a limited right that would expire, and that once copyright had expired,72 these previously-proprietary materials would be free for any member of the Australian public to use. This has not changed with the extension of the term of copyright in Australia. Under the original colonial laws, copyright reflected the English duration of life of the author plus seven years, or forty two years, for books; this period was repeated in the first national law, the Copyright Act 1905 (Cth), although members of the Parliament butted heads over the appropriate period of protection.73 The period was extended to the Berne-mandated life plus fifty years when Australia adopted the Copyright Act 1911 (Imp) as part of the Copyright Act 1912 (Cth). This period remained in place until the AUSFTA mandated a longer term of protection.

Australia also inherited from the United Kingdom the notion of legal deposit. This was included in the 1842 United Kingdom Copyright Act and incorporated into the first colonial copyright statute in Victoria. By this time, registration was not tied to protection in Australia: for example, pursuant to section 14 of the 1869 Victorian statute, copyright subsisted in “every book which shall, before or after the passing of this Act, have been or be first published in the colony of the Victoria in the lifetime of its author”. However, a registry of copyright works was kept (including books, paintings, photographs etc) and registration was a precondition to a cause of action for infringement. In addition, these early statutes required that a copy of, for example, a book that was first published in Victoria and thus received copyright protection under its individual law, had to be deposited at the Public Library of Melbourne within two months of publication (s 16). This continued under the Copyright Act 1905, where registration was still required for infringement actions; however, rather than a copy of the book being deposited to a public library, two copies of the book had to be deposited with the Registrar, upon registration (s 75). Under the Copyright Act 1912, registration was optional (s 26), although needed for an owner to avail themselves to certain remedies provided in that Act. Today, as discussed in greater detail below, we have no registration system, but legal deposit is still a requirement under our national law.

Thus the Australian public domain, as with many other national public domains, has been subject to international influences. This is not surprising given the position of Australia in the global landscape, particularly in light of our membership of the Commonwealth. This influence has been stronger at some points and weaker at others over the last 150 years. Similarly, while there have been many instances of judicial recognition that databases and compilations of information and facts are capable of attracting copyright protection, this has been matched by a consistent recognition of facts and information being part of the public domain. While the rhetoric and meaning of the Australian public domain may not be as strong as in other jurisdictions, for example, as in the United States, once you start looking it is difficult to ignore its presence.

In light of this brief historical analysis, we can turn to the elements of what now constitutes our national public domain.

4.2 Components and characteristics

What developments particular to Australia have influenced the shape of our public domain and the existence, scope and effectiveness of public rights under Australian copyright law? The elements of law, policy74 and practice that are significant in determining the nature and scope of public rights to works in Australia can be divided into six categories:

• Limits to copyright subsistence;

• Exceptions to copyright;

• Extinguishment of rights;

• Voluntary public licences (take-up and limits);

• De facto public rights; and

• Effectiveness supports and constraints.

The first three categories are mainly to do with formal copyright law, whereas the last three have more to do with practice and institutions. The elements and categories identified have similarities to those found in other countries’ public domains, but are not identical to those found in any other country. They are a unique combination that makes up Australia’s public domain. Similarly, the measures needed to maximise the benefit of this public domain to Australia – law reform, and changes to policy and practices – will be unique.

The main characteristics of each of these six categories is reviewed briefly below, emphasising to what extent the elements in that category may be distinctive of Australia’s public domain (though rarely unique). Statutory references are to the Australian Copyright Act 1968 (Cth) unless otherwise stated.

4.2.1 Limits on copyright subsistence

Australia’s public domain gets off to a very unpromising start, if we first consider its constitutional position. Court interpretations of the federal Constitution’s intellectual property clause (s 51(xviii)) have not yet placed many or clearly significant limitations on what is capable of being protected by copyright law.75 It is unlikely that section 51(xviii) could be invoked in a defensive sense, to limit extensions to the term of copyright, and it is equally unlikely to place any limitations on copyright in the name of freedom of speech. However, none of these matters are beyond dispute.76

No registration of works is necessary for protection: for example, pursuant to section 32 of the Act, copyright subsists in any unpublished literary, dramatic, musical or artistic work where the author is a qualified person, or in published works that are first published in Australia. Therefore the default position is that any works created are primarily subject to private rights, with public rights playing only a minor role. As Berne allows, Australia does have material form requirements;77 however, broad definitions of what constitutes such form means these are not significant barriers against copyright protection.

There are no significant legislative restrictions on what types of works can obtain copyright protection. Works of a “legislative, administrative and legal nature” have both Crown copyright and prerogative rights.78 Australia has implemented rental rights of various types, and moral rights. There are, of course, some limitations on those works in which copyright will be refused because they lack the minimal requirements of originality, or on public policy grounds, but they are rarely of significance. The prima facie scope of copyright’s private rights in Australia is very broad: works starting life with no copyright protection are almost non-existent.

Other factors are also relevant to the public domain’s scope. Australia’s copyright law protects compilations79 to an extent that is in many respects at least equivalent to the protections provided by the EU’s database Directive.80 Both Australian and EU protections are in contrast to the much more narrow approach taken by the Feist case in the United States.81 The breadth or restrictive nature of the principles determining the scope of implied licences to use works is another factor. It is uncertain in Australian law at present and before the High Court.82

4.2.2 Exceptions to rights

Assuming copyright does subsist in a work, what uses may the public (or specified sections of the public) make of that work without infringing?

Little can be copied without infringement: “substantial part” can mean something close to “insubstantial”,83 and this provides little scope for public rights. Similarly, Australia does not have any broad “fair use” defence,84 but only a narrow range of “fair dealing” defences of relatively fixed scope: (i) reporting news;85 (ii) criticism and review;86 (iii) research or study;87 (iv) and legal advice.88 These have been augmented recently by a further defence of uses “for the purpose of parody or satire”.89 Limited though these exceptions may be, they do provide public rights of value. The Copyright Act also contains a number of additional sections that provide exceptions to infringement of owner rights: for example, the format-shifting and time-shifting provisions that will be discussed in greater detail below. While these provisions add to what I term as “public rights”, their limited application to private purposes means that they do not add to the public domain in any significant sense.

In contrast to these narrowly-drafted “fair dealing” provisions, Australia’s public rights created by compulsory licences are more extensive than those in many jurisdictions. Australia has compulsory licences for the benefit of entertainment industries similar to many other countries. It has been argued that these compulsory licences are the principal reason for the financial success of the recorded music, radio and cable TV industries of the USA.90 Australia also has compulsory licences for educational purposes, both for reproductions of print works, and for reproductions and in-class uses of audio-visual works, which are much more extensive than are found in some other countries. These compulsory licences, particularly those in the education sector, are a distinctive part of Australian public rights, creating a closed commons91 for benefit of certain classes of users.

Lessig notes that the “permission free” resources he discusses could cost something, so long as the user had the right to buy access. In my view this is essential for a full understanding of public rights.92 On that basis, compulsory licences may be the largest and most important limitation on the right of copyright owners to unilaterally determine the conditions of use of works. If so, then it is of particular importance that these licences are administered in a manner which maximises their benefit to the Australian public

4.2.3 Extinguishment of rights

Due to the AUSFTA and implementing legislation,93 Australia has extended the copyright term in much the same way as the USA94 to the life of the author plus seventy years. This is also in line with the position in the European Union.95 Had an extended term for protection not been included in the AUSFTA the period of copyright would have remained at life of the author plus fifty years, given the Howard- Government’s earlier pledge in light of the findings of the Intellectual Property and Competition Review Committee.96 Works will therefore now enter the Australian public domain (in the narrow sense) at a slower rate than was previously the case, though no study has yet attempted to quantify this or estimate its likely effect on Australian cultural development.

It is important to note that the extension of the term of copyright applied prospectively rather than retrospectively; therefore, where a work had already entered the public domain, copyright in that work was not “revived” pursuant to the AUSFTA. Rimmer notes that:

As part of the free trade agreement with the United States, the Australian Federal Government has agreed to a prospective extension of the copyright term from life of the author plus fifty years to life of the author plus seventy years. Understandably, the Government was nervous about implementing a retrospective extension of the copyright term. Therefore, there was no restoration of copyright, for material which had already fallen into the public domain.97

This position is in contrast to the United States and the European Union, where the term was extended retrospectively.98 Thus, under Australian law, an additional period of protection was added to existing works protected under copyright: those works where the copyright had expired did not gain protection again.

A feature of the Australian situation that contributes positively to its public rights is that, although Australia has introduced moral rights, their duration is co-extensive with the economic rights of copyright.99 This is particularly important with the right of integrity, given that if this right was of longer duration than the term of the economic rights. Anyone modifying a work after it had entered the public domain (narrow sense) would still have to consider whether they might infringe the right of integrity.

The other way to “extinguish” copyright in a work is to somehow put that work into the public domain (in the narrow sense) before the copyright term has expired. As discussed later (in Part 9.3), it is questionable how this may be done under Australian law. Whether moral rights can be so extinguished is also questionable.

4.2.4 Voluntary public licences (take-up and limits)

The existence of various licences creating public rights, for voluntary use by authors within a jurisdiction, has no effect on the public domain until those licences are used, and is proportional to the extent of their uptake and the significance of the works in relation to which they are used.

Estimating the extent of use made by Australian authors, or in relation to content related to Australia, is complex.100 By mid-2006 Bildstein identified over 100,000 web pages linked to particular Creative Commons licences.101 However, he concluded that “even though Australian versions of the licences are available, the tendency for people to use the American licences is still significant.” At the least we can conclude that works under public rights licences do now constitute a significant, if modest, part of Australia’s public domain.

4.2.5 De facto public rights

These have been mentioned above in relation to the global dimension of commons. The Australian position has its own special factors in both cases, while remaining consistent with the overall global developments. There is more likelihood of search engine caching practices being held to be in breach of copyright law in Australia102 than in the USA with its broader fair use provisions, but there have been no actions in Australia to disturb the de facto searchable commons. Despite Crown Copyright applying to most primary legal materials, this has had little impact on their availability for free access.103 On the negative side, some de facto protections of privacy in the “private use” of copyright works are being eroded by the surveillance capacities of some categories of digital works.104

4.2.6 Effectiveness supports and constraints

Lessig has said that his aim was to “map a strategy for [the public domain’s] defense.”105 He talks about “crafting an ‘effective’ public domain – meaning a free space that functions as a public domain, even though the resources that constitute it are not properly within the public domain.” He could have been talking about this sixth category of elements that I have listed as contributing to Australia’s public rights in copyright. I use the same term: “effective”.

The elements in this category are related laws, or policies or practices that help or hinder the operation of elements in the first three categories (formal legal rules operating to the advantage of the public domain), or are impediments to the operation of voluntary licences of public rights. They are vital to the effective operation of public rights.

Examples of such supporting or constraining factors are that although Australia has “legal deposit” requirement for print works published in Australia, we have as yet no equivalent for digital or audio-visual works; and we have no effective registers to locate authors, or to determine whether authors have died. Another is the increasing practice of Australian academic funding bodies to require that the outputs of publicly-funded research are made available through free-access repositories. These “effectiveness” elements are discussed at many points throughout the rest of this article.

5. An agenda for reform of this public domain

5.1 A decade of net gains for proprietary rights

For over a decade, reforms to Australian copyright law have repeatedly strengthened the rights of authors and other proprietary rights-holders.106 These reforms have included the introduction of rental rights and performers’ rights (1994),107 a new right of communication to the public and new anti-circumvention (TPM) provisions (2000),108 the introduction of moral rights (2000),109 the AUSFTA parcel of protections including strengthening of the reproduction right, new performers’ rights in sound recordings, more protection for encoded broadcasts and electronic rights management information (2004),110 major extension of copyright terms and introduction of performers’ rights in sound recordings (2004),111 and further strengthening of TPM provisions (2006).112 Throughout these reforms, new enforcement measures have regularly been introduced113, and penalties steadily increased.114

During this period there have also been reforms which have strengthened some aspects of non-proprietary rights to access and use some works. These have included liberalization of parallel import restrictions, copying for people with disabilities, and compulsory licences for educational institutions (1998),115 exceptions to infringement of copyright in computer programs (1999),116 updating and extending fair dealing, computer software and other exceptions, and introducing statutory license scheme for retransmission of free to air broadcasts (2000),117 and most recently new fair dealing and other private use rights (covering parody and satire, format time and space shifting and a new “flexible dealing” exception for libraries, archives, collecting institutions, and educational institutions).118

The general trend over the previous decade has been the creation of whole new areas of proprietary rights and the carving out of parts of what had previously been in the public domain, coupled with either concurrent or subsequent “handing back” of some parts of the public domain by way of very limited and technically defined exceptions. Public rights have certainly been on the losing side in the last decade, despite the attempts made to mitigate some of the most unfair aspects of these losses.

It is not my purpose here to criticise any of these reforms, but only to point out that there has been a steady expansion of proprietary rights, often in order to comply with multilateral or bilateral obligations. If broader and stronger proprietary rights strengthen innovation, then Australia has given this impetus to innovation plenty of opportunity for over a decade. In my view it is time to give more concentrated attention to the impetus for innovation that public rights can provide.

5.2 A “Public Domain Review”

The rest of this paper details119 eight examples of where Australia’s copyright public domain needs more attention from the legislature, government policy and business and civil society initiatives if its potential contribution to Australian innovation and culture is to be maximised:

1. The scope for further exceptions to copyright;

2. Legal deposit’s role in the public domain;

3. Finding missing rights-holder (orphan works);

4. Enabling open content licensing to thrive;

5. Maximising the value of open source software;

6. Coexistence of collecting societies and public rights;

7. Re-usable government works; and

8. Public rights in publicly-funded research.

These issues are interlocking, and they are important to innovation in Australia. I suggest that they justify a comprehensive review of Australia’s copyright law (and possibly other intellectual property laws) with the question of public rights in information goods as the focus of the enquiry. Such a review has never taken place in Australia, or to my knowledge elsewhere in the world.120 Australia will benefit from a considered view that focuses on public rights and their relationships with proprietary rights, and the balance between them that will maximise the national interest. I refer to such a review in the following as a “Public Domain Review”.

6. Scope for more balance by exceptions to copyright

As we have seen, the Copyright Law Review Committee recognised “that the exclusive rights of copyright are partly defined by the exceptions, in that the rights only exist to the extent that they are not qualified by the exceptions” and considers that “the principal exceptions, such as those for fair dealing, are fundamental to defining the copyright interest.”121 The potential scope of this fundamental aspect of the public domain depends on the “three step test” for exceptions initially established by the Berne Convention. This is a very formal “black letter” place to start a consideration of reforms to Australia’s public domain, but so many aspects of expanding the public domain hinge on whether they are achievable in light of the (immovable) “three step test” established by the Berne Convention and TRIPS agreement, that it seems better to deal with it at the outset, and then to consider what scope there is for expanding the current exceptions and thus the public domain.

6.1 The 3-step test Down Under – restricting or not?

Article 9(2) of the Berne Convention states:

It shall be a matter for legislation in the countries of the Union to permit the reproduction of such works in certain special cases, provided that such reproduction does not conflict with a normal exploitation of the work and does not unreasonably prejudice the legitimate interests of the author.

This Article was included as part of the Berne Convention following the 1967 Stockholm Conference and, in the words of Ricketson and Ginsburg, “it has become the centerpiece of the exceptions regimes that have been incorporated into the TRIPs Agreement and the WCT.”122 Despite this significance, confusion still surrounds the interpretation of this Article.123

In addition to the appearance of this test in the Berne Convention and TRIPs, it was also included as part of Article 17.10 of the AUSFTA:

(a) each Party shall confine limitations or exceptions to exclusive rights to certain special cases that do not conflict with a normal exploitation of the work, performance, or phonogram, and do not unreasonably prejudice the legitimate interests of the right holder[.]

Any exception to the rights of the copyright owner introduced into Australian copyright law must therefore be in accordance with this provision, or Australia may be in breach of its obligations under Berne, TRIPS and the AUSFTA. But whether this is any significant limitation is highly contested.

The new s200AB124 sets parameters within which the “three step test” is being interpreted and applied in Australia. When it was first included in the Copyright Amendment Bill 2006, section 200AB included an exception for parody or satire. Rimmer, in the lead-up to passing of the Copyright Amendment Act 2006, described section 200AB (s200AB) as “a strange, catch-all provision” that was fundamentally “unworkable”.125 He concluded that s200AB, and the additional exceptions, as “a poor substitute to the open-ended, flexible defence of fair use.”126

Kenyon argues that s200AB has a limited role.127 He notes that s200AB merely relates to particular areas that could be subject to exceptions consistently with the “three step test”. While other exceptions should meet the “three step test”, there is no reason that the uses must fall within s200AB, nor that the exceptions should take a legislative form similar to s200AB (involving transposing wording from an international agreement into domestic legislation). S200AB will be important in Australia because it is likely to shape (and might misshape) perceptions about the options, and it may be subject to judicial interpretation which could have important implications for the statutory approach to other possible exceptions. In relation to whether exceptions comply with the “three step test”, there are major areas that require exploration about the scope of “public interest”, and room for public policy choices within the “three step test”.

A crucial point to appreciate is that existing decisions of relevant World Trade Organization Panels which have considered the “three step test” do not set definitive limits to the scope and role of public interest within the “three step test”.128 These authors argue that WTO Panel decisions are often mischaracterised as preventing recourse to public interest arguments and as setting out an interpretive approach that will simply be followed by Australian courts. It is probably more accurate to state that WTO Panels have noted ways in which “public interest” could be relevant under the “three step test”, while saying that the particular arguments did not arise on the facts in question in those disputes. This means that analyses of s200AB that attempt to formulate a set of rules (drawn, for example, from WTO Panel decisions) to apply in deciding whether a particular use falls within the section are often misleading and may well be counter-productive to the development of cultural sector practices that support Australia’s interests in innovation while also respecting the “three step test”. Thus, the issue of exceptions and the “three step test” has not in any manner been “exhausted” by the introduction of s200AB. However, whether Courts interpreting the section will take this approach is uncertain, and the issue deserves further consideration, ideally within a broad law reform setting.

Considerable recent scholarship, including Senftleben,129 Tawfik130 and Hugenholtz and Okedeji131 questions whether the “three step test” should be seen as a significant limitation on the capacity of countries to created exceptions, and instead interprets it as (in Sentfleben’s words) a “high level abstraction” intended to reconcile the many different types of exceptions that already existed when it was introduced. Patry’s interpretation is that “National governments are free to craft laws that serve their own needs and policies, including liberal limitations and exceptions.”132

A Public Domain Review would therefore need to evaluate whether the “three step test” and s200AB of the Copyright Act limit the extent to which Australia can create further exceptions/defences to copyright, or compulsory licences. It would need to ask which feasible legislative expansions, within the requirements of our international obligations, are most needed to support Australian innovation?

Some of these are discussed in later parts of this article – for example, with regard to compulsory licences for orphan works, or for government works, and limits on the operation of collecting societies. All of these would need to satisfy the “three step test”. The most likely other exceptions requiring discussion here are further exceptions needed for cultural institutions (for example archiving) or end-users.

6.2 Fair’s fair – Would “fair use” give more balance?

A common objection to the AUSFTA was that it was unbalanced: it imported the restrictive elements of USA copyright law without any of its countervailing features which give some protection to public interests and public rights in copyright. As Cutler states

One of the things we often neglect with the direct importing of legal regimes and trade agreements and international treaties, is that we do not look at what we are not importing in terms of the offsetting regimes that accompany some of these legal frameworks.

If we look at intellectual property law and copyright, while we have holus bolus with a stroke of the pen adopted the US regime under the Free Trade Agreement, what we have not imported are some of the offsetting protections.133

During the passage of the AUSFTA and subsequent implementing legislation, numerous calls were made for Australia to implement a fair-use provision. Australian Democrats Senator Ridgeway unsuccessfully proposed a “fair-use-style” section as part of the US Free Trade Implementation Bill 2004.134 Although the proposed section listed many of the existing exceptions of the Copyright Act – for example, the use of copyright works for the purpose of research or study – and it included newer exceptions (parody or transformative use) the use of the non-exhaustive term “such as” makes it much broader than the specific examples given, and there would have been considerable scope for its interpretation.

Following the introduction of the AUSFTA, the Federal Government, in mid-2005, undertook a review into whether Australia should scrap its individual fair dealing provisions and introduce a broad USA-style “fair use” provision. An Issues Paper was released in May 2005. This review was in part based upon a recommendation of the Copyright Law Review Committee, in its report into Simplification of the Copyright Act 1968:

6.10 The Committee’s recommendations in this chapter provide a means for the current fair dealing provisions to be simplified and extended into the digital environment. The Committee proposes the following changes:

▪ consolidation of the current fair dealing provisions—ss 40, 103C, 41, 103A, 42, 103B and 43(2)—in a single provision;

▪ expansion of fair dealing to an open-ended model that specifically refers to the current exclusive set of purposes— research or study (ss 40 and 103C), criticism or review (ss 41 and 103A), reporting news (ss 42 and 103B) and professional advice (s 43(2))—but is not confined to those purposes;

▪ general application of the non-exclusive set of factors provided for in s. 40(2) to all fair dealings.

The result of these three changes is to introduce a provision akin to but more precise than the open-ended US ‘fair use’ provision (s. 107 of the US Copyright Act).135

It is important to note that this was in light of considerations such as international treaty obligations, with the “three step test”, as contained in the Berne Convention, referenced early on in the report. Indeed, in justifying its reasons for this proposal, the CLRC noted that it was “consistent with Australia’s current international obligations.”136 This provision would not have been as broad as a “fair use” equivalent, but it would have been a step in this direction. This discussion was revived again in 2005. In May 2005 the Attorney-General’s Department released an Issues Paper outlining possible changes to our national copyright legislation that would bring our law into line with the ongoing digital revolution.137 One of the main aims of the review was to construct a way for legalising time-shifting and format-shifting, both of which were not permitted under Australian law, despite technologies openly available for undertaking such practices in relation to copyrighted content. Australia’s international obligations were noted in the course of the Issues Paper, including issues with the “three step test”. The position in the United States was also the subject of significant attention, with the Attorney-General’s Department raising this question for consideration: “Should the Copyright Act be amended to replace the present fair dealing exceptions with a model that resembles the open-ended fair use exception in United States copyright law?”138 The Issues Paper further outlined the options for reform, listing four possibilities:

Option 1 – consolidate the fair dealing exceptions in a single open-ended provision;139

Option 2 – retain the current fair dealing provisions and add an open-ended fair use exception;140

Option 3 – retain current fair dealing exceptions and add further specific exceptions;141

Option 4 – retain current fair dealing exceptions and add a statutory licence that permits private copying of copyright material.142

From a public domain perspective, the adoption of Options 1 or 2 would have been preferable. However, it was Option 3 that the Howard Government ultimately adopted, enacting a range of specific exceptions as part of the Copyright Amendment Act 2006. These included the numerous time-shifting and format-shifting provisions enumerated above, in addition to the inclusion of section 200AB. Although these specific exceptions created some additional public rights, it seems that the Howard Government could have gone further in protecting the interests of the public.

Weatherall notes that, contrary to the views of Australia’s Copyright Law Review Committee and the assumptions of the authors of the Issues Paper, some academic opinion cautions other countries against following the example of the United States:

There is a respectable, although by no means universally held, opinion that the open-ended fair use exception contravenes the three-step test even in the US: Ruth Okediji, “Toward an International Fair Use Doctrine” (2000) 39 Columbia Journal of Transnational Law 75. There is an equally respectable view that the fair use exception in the US may not contravene the three-step test, because the case law provides sufficient “certainty” and specificity, but that the attempt to introduce a similar exception now in another country, without its elaboration in case law, would contravene the three-step test: see Burrell and Coleman. [Robert Burrell and Alison Coleman, Copyright Exceptions: The Digital Impact (2005)]143

However, there is a strong contrary stream of academic option mentioned above. Patry refers to

US adherence to the Berne Convention, adherence where fair use was so obviously compatible with the letter, spirit, and history of the three-test step that not a single WIPO official or foreign copyright expert ever mentioned it in the four years U.S. adherence was being debated and eagerly sought.144

Given that the United States, as a member of the Berne Convention, has a more flexible way of providing exceptions to copyright through its “fair use” exception, it is still worth re-considering Options 1 or 2, the “fair dealing options” to expand Australia’s more specific and inflexible exceptions, despite the academic cautions noted by Weatherall. The United States has never been held in breach of its obligations under Berne on the basis of its fair use exception, and it seems unlikely that it would be in future. If the wording of an Australian provision was the same in essential respects as is used by the USA,145 then perhaps it would be more difficult to argue that this was not in compliance with the three step test. This is perhaps the most far-reaching question suggested in this article, so it is best to pose it at the outset.

A Public Domain Review should consider this option again and to ask whether it could be beneficial for Australia to adopt a flexible “fair use” exception to copyright, based closely on that in force in the USA. Is there any practical likelihood of Australia being found to be in breach of its international obligations if it took this approach?

6.3 Cultural institutions and the public domain

According to a major study of the impact of copyright on the digital collections of Australian cultural institutions prior to the new s200AB, inadequate public rights were impeding innovation in Australian culture. Hudson and Kenyon146 found that:

[C]opyright has had a significant impact on digitisation practices to date, including in the selection of material to digitise and the circumstances in which it is made publicly available. This has resulted in notable differences between analogue and digital collections – what could be called a “digital skew” – and has driven the content of online exhibitions, galleries and databases.147

Copyright is of such importance to cultural institutions because “they generally do not own copyright in collection items, but routinely perform acts within the exclusive rights of the copyright owner, placing them at risk of infringing copyright.”148

Among the problems Hudson and Kenyon identified were that institutions “focus[ed] digitisation efforts on works for which copyright is easy to deal with, such as items in the public domain and those for which licensing is straightforward”;149 they had “generally conservative and underdeveloped risk management”;150 that “the libraries and archives provisions are generally not applicable for public activities, such as reproducing material for exhibitions, allowing patrons to browse collection items onsite on copy-disabled terminals, or the creation of online databases”151 and that orphan works are a major problem for cultural institutions.152

They found four main approaches to dealing with the risks of digitising collection materials.

First, institutions often rely on statutory exceptions. However, the devil is in the detail; many exceptions only permit activity in limited circumstances, and typically not where digitised material is to be made available to the public. Second, institutions report dealing with copyright through negotiating for licences and assignments. Two main difficulties arise, related to the costs of individual negotiation and the impact of orphan works. Where exception and negotiation-based approaches fail, two main options remain: avoid copyright issues through the selection of works, such as materials in the public domain; or proceed with infringing conduct under a risk management strategy. 153

Hudson and Kenyon conclude that the result is a stultification of Australian cultural practices, particularly because “institutions appear resigned to withholding digital content from public access when managing copyright becomes too difficult” and “the selection of works for public digitisation is often driven, in whole or in part, by the ease of copyright compliance.”154

There is a new flexible exception for cultural institutions and other specified users in s200AB, and section 51B allows preservation copying of significant collections by key cultural institutions.155 Hudson and Kenyon consider it is too early to tell whether they will significantly reduce these problems: “On its face, s200AB appears to have the potential to allow greater preservation activities by institutions, and permit some public activities for which licensing is not possible. However, it is an exception for which users’ level of knowledge is likely to be a major influence on its practical application (as appears to be the case for fair use).”156 A review of public rights in Australian copyright will need to review whether s200AB and section 51B are working well enough to resolve the problems identified by Hudson and Kenyon, or whether it needs to be strengthened.

7. Expanding legal deposit’s role in the public domain

7.1 Importance of legal deposit to the public domain

One requirement for the effective operation of the public domain at the expiry of copyright in a work is that there is at least one copy of the work available to the public for subsequent reproduction by anyone. In Australia this requirement is satisfied for print works by “legal deposit” requirements in federal law and that of various States,157 but they do not apply to audio-visual works (now very often digital) or texts published in digital form. So there is no guarantee that a copy of a published digital work will be in a publicly accessible repository when its copyright expires. Menell sums up the danger:

The newly developed ability to preserve knowledge electronically has an important temporal dimension. Like endangered species, many forms of human knowledge are vulnerable to extinction. Therefore, societies run the risk of losing aspects of their cultural heritage by forestalling the process of digital archiving.158

In addition, if these digital works are increasingly only accessible through access control systems or distributed with technological protection measures, it is also possible that copyright in such works may expire with no copy which can be accessed technically being available.

The Australian federal government has published a Discussion Paper for an enquiry into the extension of legal deposit to audio visual and digital text works.159 The Discussion Paper does not specifically mention the importance of legal deposit schemes to the maintenance of a healthy public domain in Australia, stressing instead that the purpose of such schemes around the world is to “preserve national heritage, and to provide the public with access to that material for research or study.”160 It refers to the by the Copyright Law Review Committee in 1959, which stressed that the purpose of legal deposit was “to build up a complete collection of Australian literature”, but made no mention of the use to which such a collection should be put, and again in 1999, when it stressed accessibility to the public for research or study purposes.161

This approach does not adequately recognise the other objective function of legal deposit schemes, to provide copies of works which may be republished, or re-used in other ways, once the work is no longer subject to copyright protection because of expiry of copyright. However, even when a work is still within the term of copyright protection, if the only publicly accessible copy of it is one which is provided by a legal deposit institution, then legal deposit is essential before even the uses which are allowed under fair dealing or compulsory licence provisions may be enjoyed. These are both “public domain functions” of legal deposit, in both the narrow and the broad usage of the term. Having a copy available so that it can be used to create new, transformative, works, is what is needed to encourage innovation. The Discussion Paper is therefore somewhat lacking in that it does not explicitly address these public domain aspects of legal deposit schemes in the questions it asks.

There are at least eight questions about the role of a legal deposit scheme which need to be asked from the public domain perspective:

(i) Will the scheme guarantee that when the copyright term expires a copy is available for anyone to reproduce, so that they can obtain a copy to further transform? The USA Library of Congress indicates that less than 20% of U.S. features films from the 1920s remain wholly intact in American archives.162 This supports not only the case for archiving, but also digital archiving. This is a question posed in the enquiry into extension of legal deposit.163 Electronic versions (free from digital locks) will be easier to preserve, reproduce and distribute. This will be especially useful when the work falls in the public domain.

(ii) After a copy is deposited what steps should be taken to ensure that the item can be found? Investment should also be made in appropriate search technology to locate archived material, digital or otherwise, decades later. Searches that can look for information about a digital work (often the case with libraries now) as well as within the content of a digital works are far more desirable, especially for research purposes. Images and videos may prove to be difficult to tackle as search possibilities are more limited in such cases.

(iii) Does a scheme provide for users to both access materials held under legal deposit during the term of copyright in a work, and to exercise any fair dealing or other public rights which exist? The Discussion Paper outlines the circumstances in which the NLA, NFSA and the Pandora Project allow access to materials while they are within the copyright term, and how this may be extended to digital and audio-visual materials,164 but the emphasis is on collection practices and access. It does not address whether, and with what guarantees, users can exercise their rights under fair dealing, compulsory licensing and other existing public rights. Care needs to be taken that the interests of authors are not unnecessarily prejudiced in furthering this aspect of the public domain uses of legal deposit works.

(iv) Should the existing legal deposit schemes be re-examined from this perspective in relation to non-digital textual works as well? Should the resulting policies be included in copyright legislation?

(v) Can legal deposit schemes also be used to assist in the resolution of the problem of orphan works and authors who cannot be located? Another public domain aspect is that such deposit requirements, because they require identification of publishers of works, can also assist in identifying copyright holders of both orphan works (during the term of copyright) and the author whose death may need to be ascertained (for works out of copyright). But this would require a better searchable register of publishers and authors that legal deposit schemes provide at present.

(vi) What changes to the law (if any) would ensure that Australia is best placed to participate in any consolidation of digital libraries in the Asia-Pacific region? The European Digital Library Project (funded by the European Commission), completed in February 2008, integrated bibliographic catalogs and digital collections of various European National Libraries to create “The European Library.”165 Consolidation of digital collections is an extremely useful exercise helping to reduce duplication of archival work (and associated expense) as well as increasing the pool of resources available and accessible. Such a scheme may also help to resolve the problem of orphan works and locating authors. It may be useful to consider such a digital library project for the Asia-Pacific region.

(vii) Will the scheme guarantee that, when the copyright term expires, there will be be no impediment to access or reproduction because of the use of technological protection measures (TPMs) in relation to the work? Fitzgerald, Coates and Kiel-Chisholm consider that it is uncertain whether Article 17.4.7 of the AUSFTA prohibits exceptions to sections 116A0 and 116AP to allow depository institutions to effectively obtain and create devices to ensure that materials lodged under legal deposit schemes may be accessed.166 Alternatively, they suggest, should deposit be required of copies which are not protected by technological protection measures?

The eighth question, concerning the role that legal deposit should play in relation to audio-visual works, requires more consideration.

7.2 Expansion of legal deposit to audio-visual and digital works

The outcomes of the Attorney-General’s Department / DCITA review of Legal Deposit’s extension to audio visual and digital works is very important to the health of Australia’s public domain. The main recommendations that I and others have made to the review167 can be summarised as follows:

• There should be two criteria for inclusion of audio-visual and electronic materials in legal deposit:
  (i) For any materials (except free access materials on the Internet), if they are sold, or distributed for free, deposit by the publisher should be required under the same conditions as would make a person a “publisher” in relation to print materials. This would apply to all materials sold on CD, DVD or other medium, or delivered via the Internet by any means other than the World-wide-web).
  (ii) All materials available for free access on the Internet should be included, and provision by that means should be considered to be publication. Depository institutions should be entitled to make copies of such materials for the purposes of legal deposit, without the publisher being required to provide a copy. They should be authorised by law to ignore robot exclusion protocols168 for this purpose. However, if the publisher uses any technical means to prevent the depository institution collecting a copy of the materials, a depository institution may require deposit of copies as with (i).

• Publishers of any materials (except free access materials on the Internet) should be required to submit material in the best quality format in which it is provided to those to whom it is published. However, if the depository is unable to display the materials in the same way that these recipients can display or use the materials, it may require the publisher to either (i) provide software to allow such display or use; or (ii) provide the material in another format in which it can be read by the depository institution with no significant loss of functionality.

• Where material is available for free access on the Internet, its provision by that means will normally satisfy the legal deposit requirements, except where the depository institution cannot download the material by automated means, in which case it will be entitled to require provision of the data in accordance with the previous paragraph.

• Both electronic and print versions of a work should be required to be deposited, if both are published.

• Legal deposit should apply to broadcasts. The default position should be not that broadcasters have an obligation to deposit copies but rather that the repository has the right to collect copies. Depositories should also be given the right to require copies of broadcasts where they have not collected the broadcast when it was broadcast.

• Depository institutions should have the right, within the legislative competence of the Australian Commonwealth Parliament, to require the deposit of materials hosted outside Australia which are published on the Australian (.au) country domain or created by Australians or otherwise considered to be of cultural significance to Australia.

• Depository institutions should have the legal right to make copies of free access web sites (whether or not they are located in Australia) containing legal deposit materials for the purposes of the legal deposit scheme, and to make them searchable. The Copyright Act should confirm that depository institutions (and perhaps other search engine providers) are entitled to do this.

• In relation to deposited materials in which copyright has expired, there should be no restriction on access and reproduction. No publisher should be able to impose any such restrictions as a condition of deposit. Steps should be taken to prevent TPMs imposing such restrictions.

• In relation to materials in which copyright has not yet expired, on-site depository access should be permitted, for a single user. More liberal access should be provided to materials that are no longer commercially available. These forms of access should allow users to exercise rights of use of the materials allowed by the Copyright Act.

• Depository institutions should be part of a national scheme to identify which materials they hold that are no longer protected by copyright, and should be entitled to require information to be provided to them at the time of deposit so as to assist them to determine this.

• Consideration should be given to making searchable the full texts of all Australian legal deposit materials in which copyright has expired.

• Consideration should be given as to whether the Copyright Act should allow depository institutions to make the full texts of deposited materials searchable, provided they only provide the minimum amount of contextual information in any search results (or confirm they may do so).

7.3 Problems of identifying and quantifying Australia’s public domain

The question of how public legal deposit registers can be used to help identify works in the public domain is part of the more general question – “how do you identify the Australian public domain”, in the sense of identifying works with public rights that the searcher is able to utilise? The next section considers the complementary question, “how do you identify missing rights-holders?”169 From the perspective of encouraging innovation, how do you make these works findable so that other authors can make use of them as resources which contribute to their own work?170 It is of particular value to those who wish to innovate within their own national culture that they be able to identify and locate those works within their own national public domain: Australian creators need to be able to find Australia’s public domain.

In the past, this was only a question of being able to identify works in which the copyright term had expired. It is now a question of also being able to identify those works which are available for re-use because of public rights created by a voluntary licence (viral or non-viral), or (possibly) subject to a public domain dedication.

There are two main approaches. One is the use of Internet search engines to locate works subject to public rights licences, and are related to Australia. This approach has its difficulties but they are not insurmountable for works available online.171 The second is the use of voluntary registers or repositories of works under certain types of licences (for example, the AEShareNet register), and (if they existed) Library and other catalogs which reliably provide dates of death and allow searching by same.

The difficulties of quantifying Australian public domain content are discussed generally by Bildstein172 and subsequently in relation to the deficiencies of Google and Yahoo web searching (Yahoo having more problems) and the types of errors you get with each;173 problems in deciding why the data changed over a six-month period, again implying search engine deficiencies;174 and more generally the difficulties for quantification research caused by the lack of a cache of the web available for public research purposes.175

Global repositories of public domain content often do not give any indication of the country of origin of authors of their content. For example, SourceForge.net (one of the main software repositories) has no country-oriented metadata, and the Public Library of Science (PLos.org) containing mostly medical research and arXiv.org containing mostly physics research do not either, as is evident from their search forms.176

A Public Domain Review therefore needs to ask what public resources, if any, should be used to make Australia’s public domain, or parts of it, more easily found?

8. Finding missing rights-holders: orphan works

8.1 The problems of orphan works and missing authors/creators

“Orphan works”177 arise where the author either cannot be identified, or cannot be located, with the result that a license to use a work cannot be sought. They are a major impediment to all publishing industries and those requiring copyright permissions for performances or displays. If authors could be located, and are still alive, then they are likely to be willing to licence their works for a fee or to do so under some form of public licence (whether mere permission to use or a Creative Commons or similar licence). Both the proprietary and public aspects of copyright are harmed by the orphan works problem.

Another aspect of this problem is to determine whether an author, if identified, is dead or alive. Failure to determine this creates another category of “frozen” works, where potential public rights are ineffective because of a lack of information. Here, the main problem is the practical one of determining whether an author is dead, and if so when did they die. If it can be ascertained that they have died over seventy years ago (or over fifty years before 2005), then the work is available for public use. Otherwise, this is an orphan work problem, the issue being whether it is a live author who cannot be located or the legal successors to their copyright interest. In any event, such enquiries about the deaths of authors are part of the “reasonably diligent search” problem (discussed later).

Another related aspect is when it cannot be determined whether a work has been published during an author’s life, because if it has not, then copyright will be perpetual.178 If the work is anonymous or pseudonymous, then it necessary to know its date of publication, or it is not possible to determine when the period of copyright (s 34(1)) commences.

So there are a number of related problems here: identifying authors/creators; determining if and when they have died; locating them or their estate representatives, unless copyright in the work has expired; and determining whether a work has been published, and if so when. These are all aspects of the “orphan works problems”.

The United States Copyright Office’s study of the problem states:

Many users of copyrighted works have indicated that the risk of liability for copyright infringement, however remote, is enough to prompt them not to make use of the work. Such an outcome is not in the public interest, particularly where the copyright owner is not locatable because he no longer exists or otherwise does not care to restrain the use of his work.179

What evidence is available on the adverse effects of the orphan works problem on the Australian publishing industry, cultural institutions, and other creative industries? Hudson and Kenyon argue that for cultural institutions, traditional licensing models may fail “because of the high costs of licensing, but also because works have become “orphaned”: the copyright owner is impossible, in any practical sense, to identify or locate.”180 Passage of time, lack of meaningful attribution and breadth of items protected by copyright all add to the problems that orphan works pose to cultural institutions. They identify the problem of orphan works as one of the reasons why “the selection of works for public digitisation is often driven, in whole or in part, by the ease of copyright compliance”,181 and the content of online exhibitions is skewed as a result. Further, the time spent attempting unsuccessfully to locate copyright owners restricts the efficiency and increases the costs of the archival process generally.182

The Australian Copyright Law Review Committee, in a 1999 report, also found considerable support for an orphan works scheme, particularly from the libraries, archives and educational bodies.

Each submission in support noted that the mechanism should operate only after reasonable searches have been made and that remuneration should be paid to the copyright owner if she or he becomes known at a later date. The main arguments in favour of the development of such a mechanism were:

▪ the lack of a mechanism creates practical problems for obtaining approval for the use of copyright material and making payment for its use;

▪ the electronic delivery and creation of copyright material can be expected to result in an increase in the number of cases where the copyright owner cannot be identified or located;

▪ a lot of time and expense are expended by the users of copyright materials, trying unsuccessfully to locate copyright owners unrepresented by copyright collecting agencies, or other licensing bodies;

▪ the introduction of a mechanism would allow the freeing of access to valuable copyright material that presently cannot be accessed because to do so would constitute a breach of copyright; and

▪ the introduction of a mechanism would allow for the saving of costs associated with searching beyond what would reasonably be required to satisfy the reasonable search standard specified in the mechanism.183

Collecting societies have related problems, where they cannot locate (or cannot obtain a response from) individuals and organisations to whom they consider they should make payments. Some have online lists of persons for whom they consider they hold “undistributable funds.”184 This is an aspect of the orphan works problem.

Many submissions in response to the government’s review of the fair-use exception in 2005 voluntarily raised the issue of difficulties associated with the use of orphan works. As a result, in February 2006 the government announced that it would conduct a review of orphan works in order to address the significant problems that cultural institutions and others experience in relation to orphan works. The Australian Libraries Copyright Council (ALCC) and the Copyright in Cultural Institutions (CICI) group held a joint-seminar on orphan works with over fifty attendees from various libraries, archives, museums, galleries and other cultural institutions in May 2006.185 I am not aware of further developments since then. The Australian Digital Alliance was of the view that

The current situation where there is no provision in the Act for dealings with orphaned works, results in perpetual copyright by default. This provides a disincentive to researchers wishing to utilise a broad range of works in their endeavours, impairing or obstructing research accordingly.186

There are no easy and inexpensive ways in Australia of finding the information needed. It is difficult to determine whether and when Australian authors have died, particularly when Births Deaths & Marriages registries operate at state and territory level so there is no national register, and when a large percentage of Australia’s population, as an immigrant nation, have always been born overseas. The lack or any national ID system or residence registration system probably makes it more difficult to locate known individuals than in some countries.

There are some limited facilities in other countries which do provide publisher contact details for the representatives of some live or recently dead authors, but they are of limited use concerning Australian works. In the USA these include the WATCH service (Writers, Authors and Their Copyright Holders),187 which has some coverage of Australian authors, but searches on some well-known Australian authors produce no result; Poets & Writers’s Directory188 which lists contracts and publications for over 7,500 American authors; and the University of Idaho’s Repositories of Primary Sources,189 which is more about publications than authors but does cover Australia. The UK-based website, “New General Catalog of Old Books and Authors”190 contain an extensive catalog of old books and authors. However, the site does not offer search capability and catalogs authors by year of death or last name. Finding an author in circumstances where limited information about a publication is available to you (for example, the title of a book, or part of a name) is extremely difficult. The related Authors by Year of Death pages191 lists the authors who died in each year since 1700, to assist in finding whose copyrights expires each year, and the USA Catalog of Copyright Entries (Renewals) site192 helps with finding out whether the United States copyright of works published in the USA in 1923 through 1963 was renewed.

There do not seem to be significant Australian equivalents. The relevant author and publisher organisations in the print medium do not provide much assistance. The Australian Society of Authors website193 has an “Author Search” facility in 2007, but it only provides an alphabetic list of current ASA members who have websites, with links to them, so is of limited use to deal with the problem of finding authors or their agents generally. The Australian Copyright Council website contains information sheets about how to locate copyright owners. The Australian Copyright Council merely notes that:

Unlike the systems for trademarks, patents or designs, there is no Australian registration system for copyright, so there are no official records of ownership that you can search. For this reason, you may need to use a variety of resources when looking for copyright owners. In some cases, you may need to do some detective work.194

Copyright Agency Ltd, in an information sheet, are content with the unhelpful observation that:

Locating the rightsholder, particularly a rightsholder who is overseas, can be time-consuming. However, the fact that you have been unable to locate the rightsholder is no excuse for copying without permission.195

There are obviously significant problems here which adversely affect the operations of cultural institutions, publishers and collecting societies, among others. A first step in any Public Domain Review would be to gain a better assessment of both the significance of the problem, and why Australian institutions have not yet provided a better answer to it. What is the extent, and the economic impact, of the various problems of locating rights-holders of Australian works? The problems include identifying authors/creators of works; determining whether an author is alive or when they died; and determining who is the legal representative who may licence or otherwise authorise uses of their works (unless copyright has expired).

8.2 Options for new methods of finding authors and other creators

To what extent should obtaining copyright clearance have to involve “detective work” and be “time-consuming”? Should this be unlimited, and generate no result if the author is not found despite the searcher’s best efforts? Encouraging author associations, collecting societies, depository libraries, public domain organisations and governments to collaborate to provide a more comprehensive means of determining the status and representatives of both living and dead authors, and the publication status and date of publication of works, would be a significant step toward making Australia’s public domain – and its proprietary domain – more effective.

The USA has a voluntary register of copyright works and their owners, operated by the Copyright Office. Its Public Catalog allows searches of works registered and documents recorded by the United States Copyright Office since 1 January 1978.196 The UK Gowers Review proposed the establishment of a voluntary register of copyright works, possibly by cooperation between the Patent Office and other owners of databases.197 A Canadian Public Domain Registry is being developed in partnership by Access Copyright (The Canadian Copyright Licensing Agency), Creative Commons Canada, Creative Commons Corp. and the Wikimedia Foundation. Started in 2006, a flowchart for the project was released in April 2008.198

The questions that any Public Domain Review needs to ask fall into three clusters.

What current Australian institutions and practices provide partial solutions to the problems of identifying authors, their dates of death, their locations and their representatives? What effective schemes exist in other countries? Do they include Australian authors? Why do they not provide a better solution to the problems of orphans works and related problems?

Should Australia create a voluntary register of copyright works? What are the options for how such a register, including the organisations that should collaborate in its creation and maintenance, who should operate it, and how it would be funded? Should there be more than one such register, depending on the types of works involved? One option is that they could be coordinated by IP Australia as an online searchable public register, and involve collaboration between a number of depository institutions, organisations representing authors/creators etc.

How could such register(s) be used to help determine which works are in the public domain because copyright has expired, or otherwise indicate any public rights involved in a work? In particular, how could they be used to assist in identifying authors/creators; determining if and when they have died; locating them or their estate representatives, unless copyright in the work has expired; and determining whether a work has been published, and if so when.

8.3 A right to adopt orphan works - Alternative approaches

The United States Copyright Office held an enquiry prompted in part by the extension of the copyright term in the USA and concluded that, provided potential users first carried out a “reasonably diligent search for the copyright owner”, they should have a right to use the work, with attribution, subject only to a potential liability to pay reasonable compensation for use if the owner did subsequently emerge.199 Owners would also lose their right to injunct the publication of such derivative works (or obtain destruction of copies), provided the new author had added an original contribution to it. This USA approach can be described as a compulsory licence after reasonable enquiry, but it is not one which would involve a collecting society, and nor would it involve the payment of fees in relation to all works used.

The Gowers Review recommended that the UK government should propose an orphan works provision to the European Commission, as an amendment to the “Info-Soc Directive”.200 It further recommends that, as part of this, its Patent Office should issue guidelines as to what constitutes a “reasonable search”.201 The Copyright Board of Canada is empowered to issue licenses on behalf of copyright owners in cases where the owner cannot be located. The Australian Copyright Council recommended a similar approach in its response to the issues paper on fair use. However, the approach has been criticised due to the “unpredictability, delay and transactional expense inherent in the current system in Canada that requires a ruling from the Copyright Board.”202 The Innovation, Universities and Skills Committee in the UK has similarly proposed that the Copyright Tribunal become responsible for granting licences for the use of orphan works, but the British Academy has rejected this approach.

In its reports on simplification of the Copyright Act, the Copyright Law Review Committee did not recommend amendments to create a right to use orphan works, but did recommend further investigation of whether the Copyright Tribunal should operate some type of licensing scheme (as in Canada), or whether collecting societies should be able to represent unknown authors and licence their works in such cases (as in Scandinavia).203 In Australia, a further government enquiry was supposed to occur but does not seem to have proceeded.204

In a 2006 presentation Rimmer identifies six different structures or models that can be used to deal with the orphan works problems, which he labels: (i) Twilight Clauses; (ii) Defences; (iii) Limitation on Remedies; (iv) Registration Scheme; (v) Canadian Model; and (vi) Scandinavian Collecting Society Model.205 He notes considerable problems with some of the models. For example, “Twilight Clauses” based on the age of works, or the duration of time since publication, or the date of death of the author, will often be useless because so many of them have such little data associated with them that these dates cannot be established with any certainty. “Any system attempting to administer trust monies for all orphaned works will necessarily be complex, time consuming an unreasonably burdensome.”206 There is a risk that any scheme involving compensatory payments will “keep the orphans in the orphanage” because the uncertainty involved in the level of compensatory payments which may be required will deter use.207

Perhaps there are even more potential models than those identified by Rimmer, depending on how factors such as the following are combined: (i) whether orphan works can be used without application to some tribunal; (ii) how a test of diligent search is framed; (iii) whether the search must involve publication of a notice; (iv) whether use of the orphan work must carry a notice that it is being used, or use must be registered; (v) whether allowed use of an orphan work should be payment-free, or carry some liability for a compensatory payment; (vi) whether any liability for payment should be contingent, only arising if a rights-holder comes forward; (vii) whether any payment upon use is required; and (viii) whether collecting societies should have any role in collection and disbursement of payments.

This is one of the more important unresolved problems in relation to innovation in Australia. In assessing which option in creating some greater public rights to use orphan works, a Public Domain Review would need to start with the question of what limitations (if any) does the 3-step test impose on the types of orphan work schemes that are possible? My own inclination is to then suggest a positive answer to all of the following questions:

1. Should legislation allow use of orphan works after a “reasonably diligent search for the copyright owner” or some similar test, without any further application or bureaucratic requirements?

2. Should there also be required, as part of any diligent search, some form of public notice that the author of the orphan work was being sought?;

3. Should any uses of the orphan work be required to carry a notice disclosing this and advising whom the author or author’s representatives should contact, or some registration of its use? (The most obvious alternative is that the Copyright Tribunal or some Court or Tribunal be empowered to issue licences to use orphan works, on application and based on a set of statutory criteria.)

4. Should there be provision for reasonable remuneration to be paid to authors/creators or their estates if they subsequently come forward after becoming aware that their works have been so used? The more difficult question is then whether such remuneration should (a) only be paid at all if and when the previously missing author comes forward or (b) be required to be paid in to some trust fund as soon as the orphan work is used; or (c) be paid to some collecting society in anticipation that the owner of the orphan work might come forward.

Whatever approach is taken providing a right to use orphan works in Australia, it would work best from all perspectives if it was coupled with more effective Australian methods of locating authors, so as to make it more likely that diligent searching would locate an author who was available to be located. Another question which needs to be addressed for completeness is whether the duration of copyright in unpublished works should be limited? This is so in some countries, but not Australia, leaving unpublished works permanently outside the public domain.

9. Enabling open content licensing to thrive

Voluntary licences made by Australians, or made under Australian law, which purport to give a conditional licence of works to the public at large have become increasingly numerous since 2000, as the examples at the outset of this article illustrate. This Part discusses voluntary licences over content (texts, photos, videos etc), as distinct from software licences which are discussed in the next Part. Such voluntary licences have been used by Australian governments (for example, the New South Wales (NSW) and Northern Territory licences of legal materials to the public, or the licence to republish web materials given by the NSW Attorney-General’s Department), and by both Australian consumers and creators through Creative Commons licences. In the educational sector, the AEShareNet suite of licences developed in Australia has been used extensively. Australia’s Productivity Commission concluded that, while Universities should often commercialise intellectual property:

…[U]nder other circumstances, it is arguably more appropriate for universities to give their research away — for example, if the knowledge or technology is generally applicable to a wide range of firms and the costs of further development and replication of the resulting innovation are low. In this case, seeking to protect the IP and sell or license it delays its transfer and diffusion, potentially imposing substantial costs on firms and the wider community.208

Greater government and academic use of such licences creating public rights could stimulate innovation by creating a faster and less costly transfer of knowledge.

9.1 Evidence of use of open content licences in Australia

Some work has been done on identification and quantification of the use of open content licences in relation to Australia.209 The two most widely-used open content licences in Australia are Creative Commons licences and the AEShareNet licences, although other licences such as the GNU Free Documentation Licence (FDL) are also used. Bildstein’s figures concerning Creative Commons licences show that, as at early 2007, Australians were using both the United States210 and Australian Creative Commons licences. This was early in the life of the Australian licences, so it can be expected that, over time, there may be relatively more use of the Australian licences. As is discussed later, some major user-generated-content (UGC) websites only provide facilities to assist users to use the generic licences, not the Australian licences. Whatever relative changes in uses of the two sets of licences occur over time, it seems reasonable to expect that both sets of licences will continue to be used in relation to Australian content.

Bildstein’s analysis also indicates that usage of Creative Commons licences (of either type) which include the “non-commercial” (nc) attribute is by far the largest category of usage. The high usage of the two licences that use the viral “sharealike” (sa) attribute is the next most noticeable aspect of the figures. From the figures provided by Coates, this Australian usage seems broadly consistent with international trends. Coates sees a trend toward less restrictive licences (ie more use of sa and less use of nc) in the international figures.211 The position will become clearer as more statistics on both Australian and international use become available over time.

AEShareNet licences are the second largest source of open content licences in use in Australia. Bildstein found, as at mid-2007, nearly 4,000 instances of use of one of the the AEShareNet “instant licences” (types U, S, P and FfE), which contain “public rights elements” (the others require individual negotiation of licence conditions).212 The AEShareNet website licence database213 identified twice as many licence occurrences than can be found from searching the web (as indexed by Yahoo) for AEShareNet licences. On the AEShareNet database, the “Share and Return” licence type was by far the most used of the “instant licences”. On the web (excluding the AEShareNet website), the “Free for Education” (FfE) licence type was by far the most used licence. He conclude that it is much more common for users to publish something that is “Free for Education” on the Web than it is to register it on the AEShareNet database.

9.2 Clarifying the Australian legal status of voluntary commons licences

One of the most detailed published Australian studies of open content licensing is that carried out for the Queensland Spatial Information Office by Neale Hooper, Anne Fitzgerald and others. It explains the legal status of open content licences as follows:

Open content licences involve the granting of permission to other persons to use the copyright material in ways that fall within the bundle of exclusive rights belonging to the copyright owner. In other words, they authorise (permit) users to do certain specified acts within the scope of the bundle of rights which can be exercised exclusively by the copyright owner. Importantly, open content licences grant users rights to do acts that fall within the scope of the copyright owner’s exclusive rights and do not impose further (i.e. non-copyright related) obligations on the users of the copyright material. In this respect, open content licences differ from many traditional information licences which seek to impose, by means of a contract between the copyright owner and the recipient, additional obligations or constraints on users (e.g. limitations on re-use or confidentiality requirements).214

There is no doubt that such licences will normally be effective to those relying upon them as permissions to do acts otherwise within the exclusive rights of the copyright owner. Even those who question aspects of the enforceability of Creative Commons licences such as Foo consider that, if they are not valid as contracts, then

A more persuasive argument for the legality of open content licences that [they] effectively represent the permission required under copyright law by the owners for the user to reproduce, adapt or communicate the material without infringing the copyright in the work.215

However, some have expressed uncertainty concerning the exact status, and enforceability, of some aspects of these licences under Australian law. Foo argues that:

There is considerable debate about whether the GPL or CC licences are enforceable under Australian law. It would be difficult to conclusively state that open content licences are contracts in Australia…. The requirements for consideration and common intent do not appear to be evident in open content licences.216

The claimed lack of consideration may be difficult to dispute with those Creative Commons licences which do not contain a viral element (ie the by, by-nd, by-nc and by-nc-nd licences), but is weaker in relation to the viral or “copyleft” CC licences (ie by-sa and by-nc-sa) because it can be argued that the licensee agrees that if they modify the licensed work their modifications will be subject to the same licence. Consideration does not need to be monetary, and it does here constitute a detriment to the licensee because they are giving up some of their exclusive rights to the licensor, as well as to others. Where a licence is non-viral, the detriment is not always so obvious, although it can arise if, for example, a user has expended time and effort in modifying a work (under a by licence or by-nc licence). It will be more difficult to show detriment if a work has merely been used under a by-nd or by-nc-nd licence, but again it may not be impossible. It is difficult to estimate the practical significance of any possible problem here in the abstract: it is possibly not practically significant, but only a theoretical limit.

One possible problem with treating these licences in effect only as a defence or estoppel against copyright enforcement actions (as Foo suggests may be effective) is what happens when a previous grantor of a licence purports to revoke the licence. Perhaps, as Fitzgerald suggests,217 equitable estoppel may be available to prevent revocation of the GPL, or presumably of other viral or “copyleft” licences (such as those with Creative Commons “sharealike” attributes: by-sa and by-nc-sa)). But this would seem less likely to work to prevent revocation with those Creative Commons licences which do not contain a viral element (ie the by, by-nd, by-nc and by-nc-nd licences), despite what is said in clause 7(b) of the Australian licences purporting to make them “perpetual” (for the term of copyright in the work).

There are no significant judicial decisions concerning Creative Commons licences. As yet, only a few overseas cases have directly considered the enforceability of these licences, and none seem to be strong precedents. Foo considers that two cases have “affirmed the validity of certain CC licences.”218 First is the Dutch case concerning Curry who published photos of his family on Flickr under a CC Attribution-Noncommercial-Sharealike 2.5 Canada licence. A gossip magazine republished them, then claimed the licence was unclear. The Court ordered the magazine not to publish any of Curry’s photographs from Flickr in future. The Dutch court did uphold Curry’s claim on the basis that the licence was valid.219 However, the outcome would have been the same, whether or not the licence was valid (breach of the “noncommercial” term) or invalid (no licence, simple breach of copyright), so the case is of little significance. The second case is where a jazz club in Spain successfully resisted a claim for licence fees by the collecting society for music performances because the music concerned was licensed under a Creative Commons licence. However, it is not as simple as that. As Guadamuz explains, while the judge-magistrate did take into account the new practices of clubs playing materials that they are free to use, whether because of Creative Commons licences or other permissions, the effect of this was to nullify the presumption that works played in such clubs would be by artists represented by the collecting society.220 “With that presumption in tatters, it is the plaintiff who has the burden to prove that the music played in the locale is managed by them.”221 The individual licence here was not examined.

Hietenan, after surveying the legal status of Creative Commons licences in Scandinavian and Anglo-American legal systems, finds it difficult to come to uniform conclusions:

Giving a comprehensive answer to a question, whether Creative Commons licenses are contracts or mere permissions, is impossible. Those several legal systems that may have to deal with the licensing instruments contain different rules relating to licensing. Those rules make the legal outcomes local and the global licensing movement fractured.222

The arguments discussed here are not conclusive one way or the other, though it is notable that practical problems have not become well-known given that the licences have now been in use for over five years. They only indicate that there is some uncertainty concerning the legal basis, and limitations, of enforceability of Creative Commons licences, and that the position may differ depending on the type of licence.

Nevertheless, this seems to be an unsatisfactory situation, for two reasons. The first is that the overall purpose of the Creative Commons and similar licences is socially beneficial (particularly from the perspective of enhancing innovation): they aim to provide a uniform method, with low transaction costs, by which copyright owners can permit others to use some of their exclusive rights in relation to works. They expand the choices available to licensors, and the transfer of intellectual goods to licensees.

Second, there is substantial and growing use in Australia of the main types of such licences (those from Creative Common Australia, the Creative Commons generic licences, and those from AEShareNet), including use by our major public institutions in government, academia and the cultural sector, and by a growing proportion of Australians through their use of user-generated-content (UGC) websites.

It would seem to be clearly in the public interest to put such licences on as clear and watertight a legal footing as possible, rather than to wait for any lingering legal doubts to be resolved by court decisions, assuming it is possible for changes to the Copyright Act to achieve such an end without causing other problems.

The main question a Public Domain Review would need to ask, after making a current assessment of the principal voluntary commons licences used in relation to Australian works and the extent of their usage, is to clarify what is the legal status of these licences under Australian law and what uncertainties (if any) are there in the enforceability of these licences under Australian law by anyone who wished to enforce them. The question then follows of what amendments to the Copyright Act could make voluntary commons licences more clear in their legal effects, if any additional certainty is needed.

9.3 Empowering public domain dedications

Where authors or other copyright owners wish to voluntarily place their works completely in the public domain, it will normally be in the public interest, and in the interests of innovation, for copyright law to make it possible for them to do so. The Copyright Act could specify a formality which would be effective to achieve a public domain dedication,223 as this is a simpler question than that of the enforceability of Creative Commons licences or of the GPL.

There seem to be no policy reasons to restrain copyright owners from “donating” their works to the public domain, provided that in doing so they are properly informed of the consequences of this choice. The choices of the copyright owner are expanded, and the public benefits from an earlier transfer of intellectual goods to its use. The extension of the copyright term to the life of authors plus seventy years makes it particularly valuable to provide an alternative means by which authors can choose to voluntarily accept a shorter term of copyright. The formalities included in the Copyright Act could expressly provide that a “public domain dedication” could be expressed to only operate from some future date (for example, from the death of the author, or from fourteen years hence).

Consideration would need to be given to the options which have been used elsewhere to support public domain dedications, and whether there is any evidence that they are effective. These include the USA Creative Commons Public Domain Dedication224 and the Open Data Commons “Public Domain Dedication and Licence”.225 Creative Commons in the USA provides such a Public Domain Dedication “licence”, and so apparently has some belief in its effectiveness, though Lessig queries226 whether it could be effective under USA law.

Creative Commons Australia does not offer a Public Domain Dedication, and its validity would be uncertain under Australian copyright law. The status of such a declaration under the Berne Convention is also a matter requiring question, particularly in terms of its international enforceability. Whether it is possible for an author to renounce the right of integrity under moral rights law during the term of copyright is also uncertain under Australian law, but seems unlikely. This has implications for an author who wishes to use a voluntary licence that allows derivatives to be made of their work,227 because the copyright in the work may effectively have been converted into public rights, but if the moral rights have not, then subsequent users will still have to observe them.

A public domain dedication, if it was effective, would be much the same as an irrevocable Creative Commons “by” licence (ie one requiring attribution only, which is required by Australian moral rights law in any event). Both the Australian and the “unported” Creative Commons licences set out their revocability as follows:

Subject to the above terms and conditions, the licence granted here is perpetual (for the duration of the applicable copyright in the Work). Notwithstanding the above, Licensor reserves the right to release the Work under different licence terms or to stop distributing the Work at any time; provided, however that any such election will not serve to withdraw this Licence (or any other licence that has been, or is required to be, granted under the terms of this Licence), and this Licence will continue in full force and effect unless terminated as stated above.228

In other words, a licence cannot be terminated in relation to an existing licensee, or anyone relying on that licensee’s use of the licence, but it can be revoked in relation to any other potential future licencees. So the fact that a work was once available under a particular Creative Commons licence does not mean that it will necessarily be so available now or for the future.

This clause may be consistent with the requirements of Australian law that a “bare” licence (ie one without consideration) “may be revoked at will, or at least with reasonable notice”, but it may be that “if the bare licence has been acted upon by the licensee to the detriment of the licensee, then the copyright owner may be estopped from revoking the licence, either completely or without the granting of notice.”229 Other authors state that:

… [I]t would seem that an exclusive licence, being a licence coupled with the grant of a proprietary interest, is irrevocable except in accordance with the terms expressly set out in the grant, whereas a ‘bare’ licence is revocable at will. In between these extremes are contractual licences. Although the position is far from clear, it appears that a licensor may be restrained from revoking the licence, or at least required to give reasonable notice, where there is a clear express or implied promise not to revoke.230

This uncertainty concerning revocability of licences underlines not only that we need to ask whether the Copyright Act should provide for public domain dedications (ie permanent abandonment of all rights except attribution), but also whether the Act should underwrite the effectiveness of other more limited permanent abandonment of rights, such as “no commercial use” or “no derivatives” (at least until the copyright term expired, and such restrictions would also expire). As well as resolving what changes if any are needed to make public domain dedications effective under Australian law, a Public Domain Review be made to operate from a future date, and should moral rights be equally terminable.

10. Maximising value of open source software

10.1 Value of open source software to Australia

Many examples of the contributions of Australians to the development of open source software (sometimes called Free and/or Open Source Software or FOSS)231 and the innovation that has resulted from those contributions, have been given in the opening of this article. Many more could have been included.

The ongoing evolution of open source software, such as that licensed under the General Public Licence (GPL), has demonstrated a vitality and creativity that “closed” or proprietary software has sometimes struggled to match. Some open source software is commercial and some is not. Despite this, the community and business organisations that support open source development show a remarkable integration and respect for each other’s capacity to contribute notwithstanding the great range of sizes (from individuals to global giants) and the business models they embody (from loose groups of colleagues contributing to joint projects for a variety of motives to large profit-oriented businesses). Open source software is in many ways the most established and mature example of the operation of a commons-based production system in a global commercial environment. Dating in practical effect from the early 1990’s, it has approximately a decade more historical evidence demonstrating how it works in the real world than the more recent forms of “Open Content” licensing. As such, there is potential for fruitful comparative analysis, and there may also be further evolved indications of the sorts of problems that these paradigms experience in widespread practical adoption with real business models, problems that may warrant various forms of support or accommodation if we are to retain the maximum innovation benefit from this model. Many benefits are claimed for software developed under this model, particularly for governments seeking multiple suppliers, low initial investment, limited “lock-in”, access to free or cheap utilities, and limited or no licensing costs for large implementations. There may be a variety of impediments to the realization of these potential benefits.

This is important to Australia for a number of reasons. Due to our small market size and exposure to the products and services of almost every national and international IT industry, we have not yet developed a home-grown global scale IT company or established global industry standards, however we do have many micro- and small-to-medium IT businesses and experts, capable of contributing to the leading edge of global-scale projects.232

Such smaller contributors depend, more than larger players, on access to a range of licensing models, and low cost compatible tools to use in providing competitive services. A key feature is the ready customisability of such software and software models. This flexibility aligns well with the need to customise generic products for the special needs of specific Australian businesses or groups thereof, especially small-to-medium enterprises (SMEs), who would otherwise often be prevented by cost considerations from tailoring their tools to their requirements. Such tailoring can contribute significantly to productivity and international competitiveness by enabling local businesses to adapt software to their evolving business processes, rather than be constrained to adapt and limit those processes to match relatively inflexible generic software.

There are also substantial initiatives in, for example, the education sector.233 It increasingly benefits from low-cost compatible tools that can be easily customised. For instance, a project in the USA tertiary sector to provide open source accounting and other bespoke software products for that sector, specifically because the education sector, or particular members of it, are not “big enough” to influence existing software vendors to meet education’s needs.234

10.2 Usage of open source licences

It was until recently difficult to estimate the extent of usage of each type of open source licence, because of the lack of linkage between the code which is licenced and the licence itself, and because the usual search engines did not facilitate such searches.235 However, the release of the Google Code Search facility236 in 2007 has made one type of estimate possible. Google crawls and makes searchable as much publicly accessible source code as they can find. Google Code Search finds a total of approximately 18 million FOSS licences worldwide, of which over seventeen million are accounted for by the following five licences (in order of popularity), plus disclaimers (similar to public domain dedications): 8.9 M (gpl); 4.6 M (lgpl); 3.1 M (bsd); 0.9 M (mit); and 0.1 M (cpl).237 The remaining licences found accounted for 300 or less software items each.238 The words in parentheses are the licence names Google Code Search uses.239

The list is not very surprising, indicating that the General Public Licence (GPL),240 with nearly nine million licensing instances on the web, is by far the most popular open source licence. It is followed by the 4.6 million instances of the Lesser GPL (LGPL), the difference being that “using the Lesser GPL permits use of the [software] library in proprietary programs; using the ordinary GPL for a library makes it available only for free programs.”241 The fact that the GPL accounts for nearly 50% of all FOSS licensing worldwide makes it clear that it is a significant question whether that licence is enforceable at law, even if adherence to its norms are primarily social rather than legally-enforced. These statistics do not distinguish between GPLv2 and the recently-introduced GPLv3, which may have significant differences in terms of validity in some countries, as discussed below.

These figures are global, and it is not possible at this stage to estimate which licences are most commonly used in Australia,242 either by software developers writing new programs and choosing which licence to use, or by those using existing FOSS software, or those developers modifying existing FOSS software. Software is much more global in its use than is open content which is sometimes of much more parochial interest. There are no reasons I am aware of why the Australian percentages of use would differ markedly from the global pattern, but it is possible that there are some factors in the history of Australian software development that may indicate this.

Quantity of licence use is not the only indicator of importance. Certain licences may be relied upon in various “mission critical” aspects of the economy or of government, and the validity of those licences might not be reflected in how commonly they are used. We therefore need to ask, for which free and open source software licences is enforceability under Australian law of greatest economic and social importance? The principal answer will probably be “the GPL”, but beyond that perhaps there are some other licences in Australia which have importance more than they do elsewhere.

10.3 Validity and enforcement of FOSS licences

The open source model has in practice been widely adopted and integrated into substantial investment streams,243 which indicates some level of confidence in its validity and enforceability. Nevertheless, questions persist about the enforceability of the terms of some of the main licences, such as the General Public Licence (GPL), similar to some of the issues already discussed in relation to Creative Commons licences. There are also separate issues of the interaction of these licences with the patent system, and issues concerning the use of one “global” set of terminology, in contrast with the Creative Commons approach of “porting” licences adapted to the legal terminology of a country like Australia. As discussed in reference to open content licences, there is more likelihood of viral licences being held to involve consideration than those which are non-viral.

The relatively low level of litigation, both globally and in the Australian jurisdiction, has not yet produced any significant case law on the questions of validity and enforceability. For instance, the basis on which voluntary licences to the world are enforceable – as a contract, or as a defence to infringement, or under some other mechanism – is still untested, at least in jurisdictions like Australia. Fitzgerald and Suzor present a case for the enforceability of the GPL which is based in part on its being enforceable at law as either a licence or a contract, and in part on their being a vigorous developer community in relation to most open source software which will use informal pressure to ensure that licence terms are observed.244

The German organisation gpl-violations.org claims to have successfully taken action since 2004 in over 100 cases involving GPL violations, with total “legal success, either in-court or out of court.”245 Some GPL software developers have transferred rights in their software to gpl-violations, in order to enable it to take action.246 Some details of many of these examples of enforcement are available on their website. However, all of the enforcement actions were against German companies and under German law, and they have only in November 2007 commenced their first action outside Germany (in France).247 The most significant decision is probably that of the Frankfurt District Court judgment in the D-Link Case (Welte v Deutschland GmbH).248 The Court held the GPL valid in most respects, under German law, as the basis of deciding the case. However, the Court’s decision in the D-Link Case was that the GPL was not valid under German law in some respects, and the decision illustrates that exactly how the GPL will operate will depend on a complex interaction with the laws of each country.249 The Software Freedom Law Center in the USA similarly provides legal representation and other law-related services to protect and advance FOSS. A recent success is its settlement of a case against Verizon Communications Inc. for claimed GPL violations.250

The question “is the GPL valid in all respects under the law of country X?” does not seem to be a trivial enquiry, though there does not yet seem to be any evidence of significant problems arising in any legal systems. If this is so, there could be benefit in a detailed analysis to establish whether any legislative changes could assist in ensuring that the GPL and perhaps other FOSS licences have the greatest possible compatibility with Australian law. Another reason why this might be beneficial is that, unlike the Creative Commons licences, there is no “porting” of the GPL to the Australian legal system, it is a “one size fits all” jurisdictionally universal licence. While there were quite a few terms and concepts in GPLv2 which seemed to be primarily derived from the United States legal systems, a considerable effort was made with GPLv3 to ensure that its terminology was more neutral, based on the terms used in international agreements where possible, and able to take its meaning in some parts from the jurisdiction in which it was being applied. Both FOSS licensed under GPLv2 and FOSS licensed under GPLv3 are likely to co-exist for some time.

Since uncertainty can equate to perceptions of commercial or legal risk, even when not founded in any real or demonstrated flaw, this factor may prompt unnecessary reluctance to use the open source model in otherwise appropriate situations, limiting the development of this aspect of the public domain. It is therefore a useful question for a Public Domain Review to ask whether there any reasons to think that GPL v2 or GPL v3 (or other significant open source licences) are not enforceable under Australian law, to ask whether any unresolved questions about enforceability hinder its adoption in appropriate situations, and whether changes to the Copyright Act could remedy any potential problems?

11. Coexistence of open content and compulsory licences

“Compulsory” or “statutory” licences under Australia’s Copyright Act create rights in the public as a whole, or in particular classes of the public, to make use of otherwise proprietary content, on conditions and often for a fee set by the Copyright Tribunal or some other process. This content constitutes one of the largest and most important components of Australia’s public domain. Collecting societies administer these compulsory licences, and their practices therefore have a significant impact on the health of Australia’s public domain.251

With the many changes toward the use of open content licences in past decade, particularly with the rise in importance of the Internet, tensions have arisen. This older form of public rights, by which the public use rights of creators in relation to all their works were often administered on their behalf (often via compulsory schemes), now has to co-exist with new forms of public rights whereby creators voluntarily act through open content licences to allow some no-fee uses to be made of particular works. This raises a number of potential problems.

11.1 “Some rights reserved” from collecting societies?

The membership conditions of some Australian collecting societies have made it difficult for their members to licence any of their works under Creative Commons licences or other licences with public rights.

The Australian Performing Right Association (APRA) requires its members to assign all present and future copyrights to it, and argues that it cannot legally collect fees until this occurs. Oi explains as follows:

All APRA’s 33,000 members have to assign to APRA all their public performance rights, before APRA can collect the royalties on their behalf. Those 33,000 members include all Australian song writers and composers whose works are applied commercially. That affects a significant proportion of the creators that are already out there and working, and who may wish to participate in the Creative Commons. This is something that APRA members and anyone who potentially wants to become an APRA member will have to be aware of. They will not be in a position to use a Creative Commons licence to license their works, unless they have reached some alternative arrangement.

The wording of the Australian licence accommodates this up to a point, but there is still a danger and a risk for potential APRA members who do not realise what they are doing to potentially get themselves into trouble by trying to license out something that they may have effectively signed away to someone else. This is a follow-up area of work, and the people at APRA have been very good at giving feedback and comments on the effect and the potential interaction with Creative Commons. I look forward to working with them to develop some further commentary and to get some guidance out, and to find easier ways for creators to both work with Creative Commons and to also collect royalties via APRA. That is one area of work that needs to be done: collaboration with collecting societies in Australia and other organisations that are relevant.252

This issue resulted in Creative Commons International unsuccessfully arguing that APRA was involved in anti-competitive conduct.253 The Australian Competition and Consumer Commission (ACCC), in re-authorising APRA’s licensing arrangements, commented:

In its draft decision the ACCC expressed concerns that by generally taking exclusive assignment of its members’ rights APRA's arrangements effectively foreclose any realistic prospect of music composers and users dealing directly in most instances.

The ACCC also expressed concerns that APRA’s propensity to only offer users licences covering its entire repertoire - irrespective of the needs of the users - eliminates incentives for music composers and users to negotiate performing rights licences other than through it.

In response, APRA modified the arrangements by which it takes assignment of its members rights to make it easier for composers to negotiate rights in respect of their works directly. APRA has also indicated it is prepared to develop alternative licensing arrangements to allow direct dealing between music composers and users where the music user express such an interest.

While the ACCC is encouraged by APRA’s response, the utility of these amendments to facilitate direct dealing between music users and rights holders will still be very limited in most cases.254

APRA’s modified arrangements255 allow APRA members to require APRA to re-assign rights back to the member using two mechanisms: an opt-out process (allowing one or more of the exclusive rights to be re-assigned to them, over all of their works, and requiring up to six months notice) and a licence-back process (allowing use of a particular work on a one-off basis, in Australia only). Both procedures require the member to indemnify APRA. Neither procedure allows a member to licence an individual work under a Creative Commons licence or provide it via MySpace or iTunes or similar commercially significant channels.256 Despite these requirements, APRA claims that it “has had a policy for some time that allows you to [distribute your music files for free from your own website], while reserving your rights to obtain payment from other web uses.”257

In the Netherlands and Denmark collecting societies have during 2008 launched trials allowing creators of musical works more flexible methods of promoting their works.258 APRA states it is discussing such matters with Creative Commons Australia

The international collecting society community, through CIS AC, is now working on protocols to allow composers to use certain types of CC licences and retain their rights of payment from their collecting society. This is still in the process of drafting and negotiation, to ensure that the commercial rights administered by societies collectively are not undermined by CC licences. APRA is also currently in discussion with CC Australia in line with our international colleagues. We hope that we will be able to offer APRA members the benefit of using CC licences to self-license the use of their work for non-commercial purposes, while retaining the advantage of their APRA membership to license, monitor and collect royalties for the commercial use of their works.259

APRA’s current information page on Creative Commons260 states that APRA “has been in discussion with CC Australia for some time” about use of CC licences by its members to self-licence their works, but gives no indication of the outcome of these discussions, and is otherwise hostile to its members using Creative Commons licences.

The ACCC is now more involved in copyright matters and can become a party to hearings before the Copyright Tribunal. It has indicated that it considers arrangements which prevent “direct negotiation between copyright owners and users” can indicate anti-competitive detriment:

The ACCC considers the anti-competitive detriment from a collecting society’s arrangements will be more limited where:
• the arrangements do not prevent direct negotiation between copyright owners and users;
• the output or licensing arrangements are as unrestrictive as possible and strike an appropriate balance between facilitating the administration of copyright and allowing flexibility in licensing as appropriate;
• licence fees and conditions for use of copyright are clear and readily available to users; and
• the arrangements allow for alternative dispute resolution processes where appropriate.261

As yet these newly-enunciated principles have not been applied to the claim that some collecting society practices unduly prevent direct licensing that includes public rights elements.

Some other compulsory licences and collecting societies do not impose such restrictions on their members. VISCOPY provides “A guide and checklist for VISCOPY members on WAIVING FEES for the use of your work”.262 It makes it clear that its members may “decide to grant a free licence”, but the checklist’s fourteen questions are so detailed and suggest so many restrictions that it would never be possible for a VISCOPY member to answer “Yes” to all fourteen. In relation to a standard-form open content licence such as the most restrictive Creative Commons Licence (nc-nd) or the AEShareNet “Free for Education” licence. This may only mean that VISCOPY is stating that it is prudent for their members to obtain its advice before they consider using such licences, but Question 11 of the checklist does seem to indicate a position that its members should never allow their works to be used free for educational purposes:

11. Are your rights to receive income from the use of your work under statutory licences (eg. educational and government use licences) preserved? You can use the following wording for this purpose:

‘Nothing in this agreement will prevent the Artist/Licensor [or whatever term is used to describe you] from being entitled to receive payment for the use of his/her work under statutory licences in force in Australia or under the law of any other country, including licences for educational and government use, and the Publisher/Producer/Licensee [or whatever term is used to describe the other party] will make no claim to such payments.’263

Are collecting societies providing reasonable means by which creators can choose to use open content licensing? If they are not, then as with the creation of greater certainty for public domain dedications and commons licences, the question needs to be asked whether the Copyright Act (or the code of conduct of the collecting society) should both create a right to do so (where one is needed) and to specify an appropriate formality by which it may be exercised.

Any new review of this area would first need to consider what are the key differences in the membership conditions of Australian collecting societies insofar as their effect on their members use of voluntary commons licences are concerned. The substantive question is then whether there is a need for some or all collecting society members to have greater rights to opt-out from collecting society coverage for (a) some works or (b) some uses, and whether any such changes should be made through the Copyright Act or the code of conduct of the collecting societies?

11.2 Are collecting societies charging for the public domain?

Where creators do choose to use open content licensing, they often do not wish fees to be collected for uses of their work that fall within the licence terms. One of the clearest examples is the AEShareNet “Free for Education” licence, where it is most unlikely that any user of that licence would want Copyright Agency Limited (CAL) to be collecting fees from any educational institutions for the use of their work. The Spanish case involving a collecting society attempting to claim fees for public domain works being played in a club (discussed above) is an example of these issues starting to arise in practice, even in the courts. Are Australian collecting societies involved in such practices?

VISCOPY states that it will not collect fees on works that its members “are licensing for free”, and asks them to notify it if they are using such licences:

If you grant a free licence you need to do two things:
1. Make sure the licence agreement is in writing
2. Make sure the agreement is fair and appropriate;
and then let VISCOPY know which works and what purposes you are licensing for free, so that we will not license the use of those works for a fee – we have a form that you can use for this purpose which has been enclosed.264

VISCOPY states it will not collect licence fees on such works if they are informed by their members. The other question that has to be asked of each collecting society is whether they have adequate practices to otherwise identify works which are subject to licences which mean they should not collect fees? And do they have adequate practices to refund fees which they should not have collected?

There are grounds for concern that these issues are affecting collecting societies. It is possible that Copyright Agency Limited (CAL) may be collecting fees in relation to authors and copyright owners who do not wish to receive CAL payments. The New South Wales government has published standard letters which it recommends that its agencies send to CAL (where appropriate) stating that it does not wish CAL to collect fees in relation to any of its publications. It explains that membership of CAL by State agencies is required in order to stop “the circle of money” which it illustrates as shown at left.265 “By becoming a member of CAL and notifying them of your department’s copyright policy, it ensures that public sector information is freely available to educational institutions and other government bodies.”266 The Toolkit then provides a suggested template letter for those agencies which are not members of CAL, by which they apply for membership, agree to receive any fees already collected by CAL on their behalf, and then varies the normal terms of CAL membership by the following statements disclaiming any wish to participate in the CAL licence scheme or obtain any further fee distributions from CAL:

The Department does not wish to participate in any voluntary licence schemes. CAL is not authorised to, and should not, collect any moneys related to Department copyright material in relation to any voluntary licence scheme….

The Department recently clarified its copyright policy with respect to works owned by the Department. The new copyright policy allows the public to deal freely with all Department works, unless otherwise marked for restrictive use. This right extends to individuals, private businesses, government agencies, educational institutions and any other organisation, and includes the right to distribute, reproduce and communicate for any purpose. As such, the Department asks CAL to ensure that educational institutions, government agencies and any other licensees from which CAL collects copyright fees are not charged for the reproduction or communication of works owned by the Department.267

The Toolkit includes a similar letter for agencies that are already members of CAL.268 I am not aware if other governments have similar practices or precedents for agencies to use.

These issues are not explored fully in this article, and do not have the benefit of clarifications from CAL. However, if these practices are occurring, even if they are required by law, then this is a good reason for a review of the obligations imposed on CAL and other collecting societies in order to decide where it is no longer socially desirable that it does collect funds and to determine what steps are necessary so that it is not required to do so.

APRA’s limited procedures by which its members can regain rights to licence their works directly have been noted above. However, most organisations pay APRA a flat licence fee, and there is no provision for rebates where an organisation decides to use some material which is licensed under Creative Commons licences for commercial use.269 This would seem to have two effects: (i) the public, via licence fees, continues to pay for material for which the copyright owners do not intend there should be payment and (ii) there does not seem to be any incentive to organisations to play such material. It is reasonable to ask whether this might have significant anti-competitive results, but that will depend in part on the amount of content which is likely to be licensed in such a way.

The general question “are collecting societies charging for uses of materials in the public domain?” is shorthand for a complex set of questions that need to be answered by a Public Domain Review. The answers are likely to differ between collecting societies.

(i) Are collecting societies charging fees for works where it is inappropriate for them to do so, where such works are under open content licences, or where the copyright owner has made it clear that they do not wish such fees to be collected?

(ii) Do collecting societies have adequate practices to allow their members to inform them of works on which they do not wish fees to be collected, and to observe their members wishes?

(iii) Do collecting societies have adequate practices where non-members do not wish the collecting society to wish to collecting funds in relation to works in which they are the rights-holders?

(iv) Are collecting societies required to collect funds in relation to materials provided for free access where it is not socially desirable that they should so collect them?

(v) Do collecting societies have adequate practices to otherwise identify works which are subject to licences which mean they should not collect fees?

(vi) Do collecting societies have adequate practices to refund fees which they should not have collected?

11.3 Previous reviews do not cover these issues

The only independent review of the practices of Australian collecting societies was the Simpson Report, the Review of Australian Copyright Collecting Societies, more than a decade ago.270 That report was written before open content practices were significant, and it did not hold public hearings or call for submissions.271 There has therefore not been a study of Australian collecting society practices since the Internet became commercially significant. Recommendation 37 of the Simpson Report was that “as a matter of urgency, further study be made of the impact of new technologies on copyright collecting societies and potential new methods of collection”.

The Simpson report did make some recommendations relevant to the public domain and the rights of both the public and collecting society members, only some of which were acted upon by the Federal government (including the creation of Viscopy). The recommendations included: that collective administration should be the preferred mid-way house between the exercise of individual exclusive rights and a compulsory statutory licence where mass usage requires that the community be given access to the rights on reasonable terms. (Recommendation 2); that there be a multiplicity of societies so that individual societies can represent the disparate interests of the separate groups of rights owners. (Recommendation 3); that Declared Societies be required to allocate a specific proportion of gross distributions and undistributable funds, to cultural purposes. (Recommendation 12); that Government should not consider the imposition of statutory licences where commercial voluntary licences, collectively administered, are effective. (Recommendation 13); that there be no statutory licence introduced to grant access to copyright material for the purpose of multi-media exploitations (Recommendation 16); and that Qualified Societies retain the protection of section 51(3) of the Trade Practices Act 1974 (Cth) (Recommendation 26).

Seen in hindsight more than a decade later, these recommendations do not seem to have a great deal to offer the public domain or the stimulation of innovation. Recommendation 3 was apparently adopted as we now have more collecting societies than in 1995. Coupled with the rejection of any statutory licence to allow exploitation of multi-media works, innovators are faced with dealing with a multiplicity of organisations representing rights-holders, and with no option but to negotiate fragmented rights clearances.

Since 1995 there have been no major reviews of the practices of collecting societies.272 There is a periodic review of the Code of Conduct for Copyright Collecting Societies, which is mainly concerned with the handling of complaints received by the societies. The fifth and most recent review concluded that there was a very high degree of compliance with, and commitment to, the Code by all of the collecting societies.273

Collecting societies are now required to devote some portion of undistributable funds to cultural purposes. Undistributable funds’, where authors really cannot be found despite diligent efforts,274 are something like a collecting society tax on orphan works, works for which the society cannot find an author to pay. Now that these “cultural purposes” funds have been in operation for some years, it may be time to examine whether the purposes for which funds have been allocated have been appropriate, and whether their administration has met the desired levels of transparency and accountability.275 If CAL and other collecting societies collaborated in the development of a national facility to find authors (or their representatives if deceased, as prevously discussed), then this facility could also be used to assist in locating authors to which undistributed funds were owed. CAL’s counterpart in Canada is already collaborating with Creative Commons Canada on the development of such a national facility.

Two other recommendations which were not acted upon were that that there be established the position of Ombudsman of Copyright Collecting Societies (Recommendation 19) and that the Tribunal have the right to review determinations of the Ombudsman (Recommendation 22). There might now be more justification for such proposals, in an environment where clashes of interests between collecting societies and both their members, and sections of the public, seems more likely. However, it could also be the case that the current Code Review process, discussed above, adequately serves this function (even though it does not adjudicate in disputes).

All of these aspects of the practices of collecting societies affecting the public domain needs consideration by a more broadly-based Public Domain Review.

12. Re-usable government works

The Berne Convention leaves it to the individual jurisdiction to determine whether “official texts of a legislative, administrative, and legal nature” are to be protected under national copyright law.276 How individual countries have addressed these issues varies because this Article “permits a high degree of flexibility, enabling members countries to give effect to their differing views of the public interest.”277

12.1 International momentum toward re-use

At one end of the spectrum is the United States, where a range of government works, including legislation and case law, attract no protection. In the middle are jurisdictions that may provide some protection for legislation or case law, or other government-created works. At the other end of the spectrum all government-produced materials, including primary legal materials, are protected by copyright, or some other type of proprietary right. As a broad generalisation, Australia falls into this category.

Crown copyright places government-generated works out of the public domain and into the hands of the Crown as copyright holder.278 Permission for re-use of information must then be sought from the relevant office in each jurisdiction.279 There are no consistent policies governing the re-use of government materials across Australia or an equivalent to the general availability of United States government works for re-use or the European Union’s Directive on the re-use of public sector information.280 The latter was introduced in 2003 with a view to expanding the use of public sector-generated material in the Members States of the EU, an industry with a value believed to be up to EUR 48 billion.281 The “General principle” of the Directive is set out in Article 3, which states:

Member States shall ensure that, where the re-use of documents held by public sector bodies is allowed, these documents shall be re-usable for commercial or non-commercial purposes in accordance with the conditions set out in Chapters III and IV. Where possible, documents shall be made available through electronic means.

Additional Articles provide the mechanics of the Directive, including the usage of licences in compliance with the Directive. It is understood that in 2006 the European Commission undertook action against five of its member states for failure to comply with the terms of the Directive, indicating the seriousness of this Directive within the EU.282 Strong though it is, the Directive leaves it to member states to decide “where the re-use of documents held by public sector bodies is allowed”.

Although the OECD has not yet adopted a general policy concerning re-use of government information, an OECD Working Party on the Information Economy Workshop on Public Sector Information (PSI) received an expert recommendation of a set of principles including that “PSI holdings should be exempted from IPR and also copyright and data-base protection regimes.”283

12.2 Australian developments

There are no consistent policies governing the re-use of government materials across Australia or an equivalent to the general availability of United States government works for re-use or the European Union’s “Re-Use Directive”. The Australian Federal government has a Statement of IP Principles for Australian Government agencies, which refers to the “desirability” of PSI being available to create commercial opportunities, and that non-exclusive provision is preferable.

There are some significant developments at state level in Australia. The Queensland Spatial Information Council has recommended that Queensland state government agencies move to an information licensing framework for government information, based on Creative Commons licences, where no issues of privacy, confidentiality or other legal or policy constraints apply.284 Pilot agencies were identified for implementation of the Government Information Licensing Framework (GILF). A toolkit has been developed for pilot projects, and agencies identified to carry them out in the next stage.285

The New South Wales Government has also moved towards a more permissive stance on the reproduction of Crown copyright-protected materials, although not quite as extensive in development as the Queensland position. In 2006 the New South Wales Attorney-General’s Department developed the Copyright Management Toolkit, providing guidance to government agencies on copyright issues ranging from website copyright notices to interaction with collecting societies such as CAL.286 The templates provided in the toolkit were permissive in nature, and it was noted that such an open policy would be appropriate for the majority of government-produced materials. A circular was released by the NSW Government Department of Premier and Cabinet encouraging agencies to adopt the policies outlined in the toolkit.287 A study of the intellectual property notices provide on NSW government websites found very wide disparities in approaches.288

NSW statistics show that 52% of NSW government materials provided via its websites cannot be used for any purposes, and a further 45% only for non-commercial use. Only 3% can be used for some commercial purposes.289

12.3 The anaemic Crown Copyright review

Compared with these developments, the report of the Copyright Law Review Committee on Crown Copyright290 is surprisingly timid. It recommended a few reforms to Crown copyright but the federal government has not yet acted upon them, despite three years having now elapsed since the completion of the report. The CLRC’s recommendations were extremely limited, being primarily to replace Crown copyright with a clarified position of the rights of the Crown as employer over works made in the course of Crown employees’ duties (Recommendations 1 and 3), plus the abolition of copyright (and any accompanying right, for example, the Crown prerogative) in legislation, case law and similar works at the Federal, State and Territory levels (Recommendation 4). No general licensing scheme for government works was proposed, nor any more general abolitions of Crown copyright. The CLRC’s terms of reference were extremely broad, and included an explicit requirement for it to consider the rationale for government ownership of copyright material. Despite this, the CLRC does not seem to have seriously considered (or given reasons for rejecting) any of the alternative ways by which more substantial changes could be made to put Crown materials in the public domain.

The alternatives that could have been considered would at least include (a) complete abolition of Crown copyright (and reliance on other types of law to protect public interests); (b) an attempt to categorise what content should be subject to Crown copyright and what should be in the public domain; (c) an opt-in scheme by which Crown-generated content was in the public domain unless government opted to claim copyright over it by some declaratory mechanism; (d) a requirement on governments to licence to the public the use of government information (or declare it to be in the public domain), generally at no cost; or (e) a drastically shortened term of Crown copyright (instead of mere acceptance of a “publication plus 50 year” status quo; it should also be noted that unpublished Crown works will never enter the public domain). In effect, there has not yet been a comprehensive consideration of how a public sector public domain in Australia could stimulate innovation – quite clearly recognised in the European Union Directive - and serve the public interest in other ways. The CLRC’s report was a missed opportunity rather than a reason to accept the Crown copyright status quo.

In its 2007 report the Productivity Commission explains that the federal government has in recent years moved from cost recovery (based on Crown copyright) to free access with some important data:

In 2002, the Australian Government agreed in principle to the Productivity Commission’s review of cost recovery (PC 2001a) to funding the ‘basic information product set’ of its agencies from taxation revenue (Minchin 2002, attachment 1). Basic information products are determined in reference to ‘public good characteristics’, significant positive spillovers, and other Government policy reasons. Subsequently issued cost recovery guidelines contain advice to agencies on determining basic information products (Australian Government 2005a). Agencies such as the ABS, ABARE and the Australian Institute of Health and Welfare now provide data and information online free of charge to users.291 (citations omitted)

However, the Productivity Commission’s report does not take up the general question of what approach to ownership of government-produced information goods would best serve Australia’s productivity or capacity for innovation.

Cutler is more forthright in arguing that “the failure of government to address the issue of Crown copyright is extraordinary”292 and that:

… a change in policy so that governments put the IP assets they develop or control – our assets – back into the public domain is one of the crucial things that could make an enormous difference to not only access to content but also industry development in Australia.293

Public pressure for open PSI has been limited in Australia, but is growing. The Conference Report of the Australian National Summit on Open Access to Public Sector Information,294 held at QUT in 2007 includes in its “Stanley Declaration”

The adoption and implementation by governments of an open access policy to public sector information (PSI) will ensure the greatest public benefit is derived from the increased use of information created, collected, maintained, used, shared, and disseminated by and for all governments in Australia.295

12.4 A re-usable public information seal?

Even if standards, licences and practices are developed separately by each government in Australia for the re-use of public information, it will still be valuable to attempt to obtain some level of uniformity across Australia, particularly for the purpose of communicating to the public which information may be re-used without seeking specific consent. Should a new “Re-usable government information” licence be developed, with the aim of obtaining consistent usage across Australian governments? If it is not possible to have a uniform licence across all Australian governments, should there be some agreed standard for “re-usable government information” in Australia, perhaps with a distinctive stamp or logo to indicate this? The logo opposite was developed by AustLII as a demonstration, with the intention that when this “seal” was placed on a document, it would be linked to a web page setting out a minimum set of conditions to which the licence complied, allowing reproduction of the information. Would it mean reproduction was allowed for any purpose or only for non-commercial purposes? It could also be suggested that the minimum conditions would include restrictions such as, “no distortion of the information”, “no use of Crown indicia”, and “no misrepresentation that the information is ‘official’.” However, all of these restrictions may be imposed by existing laws (differing between jurisdictions), and do not need to be included in any conditions.

Such an approach would allow governments to choose to use Creative Commons licences, or public domain dedications, or their own licences or disclaimers as they saw fit, but if their methods met the minimum “re-usable” requirement, they could carry this seal, and the public would be much the wiser.

12.5 Options and questions

So the options in relation to re-use of government works include the complete or partial removal of Crown copyright, and adoption of government policies supporting the use of public rights licences (to grant re-use rights in areas where Crown copyright is retained). As yet, Australia has no coherent approach. However, the adoption of a more permissive approach to the management and licensing of public sector information could substantially contribute to Australian innovation. A Public Domain Review needs to re-consider this issue, and ask at least the following questions:

1. What benefits would arise from broader rights of re-use of government-created works in Australia? Which benefits cannot be achieved merely by providing public rights of access?

2. What is becoming international best practice in providing for re-use of government information?

3. Which recommendations by the Copyright Law Review Committee should be supported?

4. For which categories of “government works” could copyright be abolished, and under what conditions (if any)?

5. What mechanism(s) for allowing greater use of government information are most desirable?

6. Should there be a legislative requirement of government licensing of certain categories of government-created works, and if so which?

7. Should a new “Re-usable government information” licence be developed, with the aim of obtaining consistent usage across Australian governments?

8. If it is not possible to have a uniform licence across Australian governments, should there be some agreed standard for “Re-usable government information” in Australia, perhaps with a distinctive stamp or logo to indicate this?

13. Public rights in publicly-funded research

Research is at the heart of innovation. There are at least three aspects of the rights that the public (including other researchers) should have to access and to use publicly funded research: access to research outputs such as articles in journals; the ability of Universities to communicate the theses and dissertations of its students; and access to and use of the data underlying research outputs.

13.1 Public access to research outputs

Cutler sees a serious lack of balance in Australia and elsewhere in the “mindless obsession with the notion that success is getting intellectual property out into a spin-off company as quickly as possible”.296 Instead, he argues:

The more rapid the technology diffusion, the more rapid the take-up, the greater the externalities that arise from the wide-ranging penetration of new ideas and know-how. But that notion of realising the community benefits of the externalities is completely at odds with the notion of expropriating public sector funded knowledge into the micro-economic level of the firm and start-ups and so forth.297

Free access to publicly-funded research outputs is an important part of any broad notion of public rights in works to support the type of technology diffusion advocated here. The Australian Research Council’s current policy has been moving in that direction, with he funding rules for schemes it operates being amended to encourage researchers to consider the benefits of depositing their data and any publications arising from research projects in appropriate repositories. The rules for its funding schemes298 now contain statements such as:

The ARC therefore encourages researchers to consider the benefits of depositing their data and any publications arising from a research project in an appropriate subject and/or institutional repository wherever such a repository is available to the researcher(s). If a researcher is not intending to deposit the data from a project in a repository within a six-month period, he/she should include the reasons in the project’s Final Report. Any research outputs that have been or will be deposited in appropriate repositories should be identified in the Final Report.

The Productivity Commission notes that “[t]he action of the ARC and the NHMRC is consistent with an international trend. For example, funding agencies in the United States and the United Kingdom have, in recent years, encouraged access to the results of publicly-funded research, although there is variation in approaches.”299

If the ARC started to require that Australian academics report whether they have provided free access to their research outputs,300 this would increase compliance with this otherwise voluntary scheme, because Final Reports to the ARC may affect eligibility for future grants. Such a requirement would also encourage many journals in which academics publish to make their content available for free access, so that they could advise prospective authors that publication with them will fully discharge their obligations to the ARC. With research outputs no time frame is suggested by the ARC, so ARC-compliant journals could still publish for free on the Internet one or two issues behind print publication if they wished. While it is possible to capitalise on and reinforce301 the ARC voluntary approach, it is questionable what impact it will have.

The Productivity Commission received numerous submissions on the role that funding bodies should play in relation to dissemination of research outputs.302 It concluded that a stronger approach than that proposed by the ARC and NHMRC was needed:

The Commission continues to hold the view that funding agencies should take an active role in promoting open access to the results of the research they fund, including data and research papers. Although the ARC and NHMRC’s recent announcement of promoting voluntary access is to be commended, the Commission considers that the progressive introduction of a mandatory requirement would better meet the aim of free and public access to publicly-funded research results. US experience suggests that voluntary compliance by authors would be very low.303

Given overseas developments, and the Productivity Commission’s views, it seems likely that some form of compulsory open access to academic outputs in Australia will soon develop.

The Productivity Commission considers various mechanisms through which the costs of such open access publishing can be met, but does not itself provide any precise meaning as to what “open access” should mean in this context:

Funding agencies need not prescribe the form that open access should take, whether through a conventional scientific journal, an open access journal or a repository. But they would need to provide guidance on what forms of publishing would satisfy its open access requirement. This could link to the work currently done by the Australian Government on the Accessibility Framework, under Systemic Infrastructure Initiatives and under NCRIS.304

The Productivity Commission, in referring so generally to “open access” does not deal with some of the main questions that need to be considered. At its most limited, “open access” can simply mean free access to content otherwise only available from commercial sources. Mere free access does not allow subsequent users/innovators to do anything beyond what ‘fair dealing’ copyright principles allow. Alternative approaches to “open access” allow re-use of the content, either in limited forms (eg no commercial re-use) or unlimited re-use. In the case of publicly-funded academic outputs, at least some more extensive rights to quote from works or build upon them than is provided by fair dealing might be appropriate. These alternatives would no doubt still preserve the author’s rights of attribution and other moral rights.

The fundamental question that any Public Domain Review needs to answer is obviously whether the Australian Research Council and other Australian government bodies providing research grants from public funds should require all research outputs to be available in a free access repository within a period of time, as a condition of the grant? Beyond that, however, are a series of more complex questions:

1. How are such publicly-funded research outputs which come within the government policy requirement to be identified by potential third party users? Are individual items to be so marked, and/or are particular repositories to be so certified?

2. Should such research outputs be required to be licensed to the public under some minimum-definition (or higher) public rights licence, so as to allow re-use by third parties, not merely free access and existing fair use rights?

3. Alternatively, should there be more liberal “fair use” conditions in the Copyright Act applying to any research outputs which come within government policy requirement?

4. Are any further Copyright Act protections (such as “safe harbour” schemes) needed for University and other repositories which implement public policies in relation to open access to research outputs?

5. Should any of these provisions be retrospective, so that fine distinctions do not have to be drawn between what uses can be made of research outputs depending on the date of the research?

Unless some attention is given to providing appropriate protection to both those who operate repositories, and to the users of research outputs, merely requiring deposit may achieve little.

13.2 Electronic dissemination of postgraduate research

Issues similar to those discussed above in relation to general research outputs also apply to postgraduate research. There are copyright problems concerning dissemination of postgraduate theses which would benefit from support by legislation or by government policies or practices.305 The OAK Law Project argues that there has never been any questioning of the practices of universities thesis copying services, which enable others to read unpublished theses. However, with electronic theses and dissertations (ETDs), they find after a detailed analysis that there is no clear way by which universities can be confident that fair dealing or other provisions will protect them against liability for electronic dissemination of theses.306 They concluded that there was a need for copyright reforms to ensure that those disseminating research such as repositories could rely upon the same “fair dealing” defences as those who did the research (eg defences for private study or research).307

Given that postgraduate research is a major source of innovative ideas, impediments to its publication need to be examined carefully in any review of public rights in copyright. However, care would need to be taken to ensure that any such exception was not so broad as to allow abuse of the interests of third parties whose works were quoted by the postgraduate student. A Public Domain Review therefore needs to ask whether impediments to publication of postgraduate research by universities need to be removed, while at the same time protecting the interests of third parties whose works are quoted in postgraduate dissertations.

13.3 Open access to research data

Since 1996 there have been many declarations supporting sharing of and/or free access to research data (as distinct from outputs of research, discussed in the previous two sections). Fitzgerald and Pappalardo identify and discuss the following relevant policy statements:

International organisations – among these policy statements are the Bermuda Principles, the Budapest Open Access Initiative, the Berlin Declaration on Open Access to Knowledge in the Science and the Humanities, the World Summit on the Information Society (WSIS) Declaration of Principles, and the Organisation for Economic Co-operation and Development (OECD) Declaration of Access to Research Data from Public Funding;

Governments and public sector research funding bodies – including the National Institutes of Health (NIH) Data Sharing Policy, the European Union’s Directive on the re-use of public sector information, the Australian Research Council (ARC) and the National Health and Medical Research Council (NHRMC) funding policies, and the Office of Spatial Data Management’s (OSDM) Spatial Data Access and Pricing Policy; and

Private sector organisations – such as the Wellcome Trust Position Statement.308

Of the international declarations, the OECD developments are probably of greatest relevance to Australia, even though they do little more than state a common commitment to open access to research data.309 Steps consistent with these OECD recommendations are being implemented in Australia, but it is contentious whether the implementation is forceful enough.

In relation to academic research data resulting from publicly funded research, Fitzgerald and Pappalardo summarise the new position taken by the Australian Research Council (ARC) since 2007 as a policy which “does not expressly state that research outputs must be deposited in an open access repository within a six-month period, but simply encourages researchers to do so.”310 But the obligation to justify non-compliance in the project’s Final Report means, they argue “that, at least as far as research data is concerned, it is little short of a direct mandate.”311 They note that the NHMRC (Australia’s main medical research funding body) does not go quite as far as to make such explanations mandatory.312

The Productivity Commission’s draft research report on Public Support for Science and Innovation suggested (in draft finding 5.1) that “[p]ublished papers and data from ARC and NHMRC-funded projects should be freely and publicly available.” It adhered to this view in its final report, discussed earlier, stressing that this should be a mandatory requirement. As with research outputs, a Public Domain Review should first ask whether repository publication of research data should be mandatory, and then proceed to the more complex set of subsidiary questions.

14. Holistic (re-)views of national public domains

This article has raised many questions313 which need to be answered before we sufficiently understand Australia’s copyright public domain, and the role it can play in stimulating innovation, supporting Australian culture and meeting the public interest. A series of themes recur in these questions, including what latitude the Berne Convention and other international agreements allow for further exceptions to copyright protection; the roles of a copyright register or deposit schemes in giving practical support for the public domain; the various roles that voluntary licensing can play; whether the legal foundations for such licences in Australian law need strengthening; and the need for further copyright protections for institutions, including Universities, carrying out government policies in increasing access to research and research data.

This article has argued that, in Australia, the significance of these issues justifies a law reform review that has as its focus the copyright public domain as a whole, rather than small aspects of the public domain seen through a focus on some other copyright or public policy issue.

The Australian Law Reform Commission may be best placed to undertake such a large-scale review, which will require consultation with many stakeholders over a considerable period of time. The Copyright Law Review Committee no longer exists, and I am not suggesting it should it be revived for this purpose. Nor is an internal review by the Attorney-General’s Department appropriate, because it is responsible for the day-to-day implementation of government policy in this area. Weatherall argues that the history of changes to the Copyright Act in recent years show a decline in the significance of reviews by external bodies and an increase in the involvement of the executive branch of government in the ongoing operation of the Act.314 Independence from the operation of the “copyright industries”, and lack of any long history of engagement with its stakeholders, may be advantages in reviewing copyright law from such a different perspective. The ALRC has experience and a successful track record in conducting major and innovative reviews of intellectual property issues, having previously completed an internationally renowned reference on gene patenting and on protection of human genetic information.

Similar issues and themes may arise in relation to the public domain of any country, but we can be sure that the particular copyright laws, and the particular institutional arrangements, to be found in any country will make the precise collection of issues that need to be addressed distinctive for each country. International copyright agreements, and the global impact of the Internet, mean that some of the issues have the same foundation everywhere, but their precise formulation is a product of local institutions and national copyright laws.

315

---