This essay discusses the application of Australian copyright law to the phenomenon of hyperlinks which has become ubiquitous on World Wide Web sites on the Internet. In particular, I consider how the copyright law applies to:
I begin with brief background explanations of the Internet, the World Wide Web, the technological infrastructure within which hyperlinks operate and copyright, before proceeding to consider the application of copyright law with a view to determining whether it can cope with hyperlinks.
The Internet had its genesis in the middle of the Cold War when the US Department of Defense sought the development of a network of geographically dispersed computers which could continue to function when there were a number of damaged computers or telephone lines linking them. Rand Corporation, a research and development think tank for public policy set up with the encouragement of the US Air Force, came up with the idea of a packet- switched network.
Packet switching involves breaking data transmissions into discrete packets of data, each of which contains information about its destination. As each packet contains the destination information, it does not matter what route it takes to get to its destination. This feature of packet-switched networks achieved the goal of data transmissions navigating a partially damaged computer network. Once the packets reach their destination, they are reassembled into the original data transmission.
Although the US Advanced Research Projects Agency (ARPA) set up a packet- switched computer network (ARPAnet) in 1969, it was not until 1982 that a new protocol -- Transmission Control Protocol/Internet Protocol (TCP/IP) -- was developed which enabled the interconnection of not just computers, but disparate and geographically remote networks of computers.
Originally confined to the US military and the research departments of US educational institutions, the growing presence of corporate institutions on the Internet started a trend which opened up the Internet to more widespread membership. Today, the Internet is an amorphous collection of thousands of computer networks all over the world which are interconnected in real time with TCP/IP over satellites, dedicated lines and normal telephone lines. Perhaps more significant than the millions of individual computer systems connected to the InternetFN1 are the millions of people who use those computer systems.FN2
Before the advent of the World Wide Web (WWW), the main use of the Internet was for electronic mail (email), the exchange of data files using the File Transfer Protocol (FTP), the retrieval of information using a menu-driven hierarchical text interface (Gopher) and topical discussions in Usenet newsgroups.
In 1989 Tim Berners-Lee, a researcher at the European Laboratory for Particle Physics in Switzerland (CERN), wrote a proposalFN3 for the management of information at CERN using a linked, non-linear text system known as hypertext. His proposal discussed the shortcomings of existing systems which relied on a tree-like hierarchical structure (eg the Gopher menu system). It is from that proposal that a text-only version of the WWW was born. From one text-only web site in 1991, the web has experienced exponential growth and now numbers millions of multimedia web sitesFN4 which comprise hundreds of millions of virtual pages.FN5
Using the HyperText Transfer Protocol (HTTP) over TCP/IP, a computer running WWW client software known as a web browser fetches a "page" from another computer running WWW server software. A specific page is fetched by referring to its Uniform Resource Locator (URL) which is effectively the unique address of the particular page. For example, the URL:
http://www.server.com.au/test.html
indicates a request for the HTTP transfer of the virtual page described in the file "test.html" which is stored on the WWW server "www.server.com.au". HyperText Markup Language, web browsers and hyperlinks The lingua franca of the WWW is the HyperText Markup Language (HTML). The logical structure of each element of a virtual page (eg headings, paragraphs and hyperlinks) is described by marking the text up using HTML element tags. Thus, a first level or major heading is marked up by enclosing the heading text between an opening heading element tag "<H1>" and a closing first level heading element tag "</H1>". A web browser uses the information supplied by the tags to format headings and other elements on a user's screen in a particular way depending on the technical capabilities of the web browser software, computer system and display screen.
Hyperlinks allow someone using a web browser to select some text or image to retrieve a particular virtual page, usually by clicking on the text or image with a mouse pointer. This is obviously a much easier way of navigating the WWW than having to carefully type in what are otherwise obtuse and difficult to remember URLs.
A textual hyperlink is created by using the following HTML markup:
<A HREF="http://www.server.de/test.html">Click here for German Test page</A>.
This markup has the effect of highlightingFN6 the text "Click here for German Test page" when displayed by a web browser. When a user selects this text, the browser uses the HTTP protocol to fetch a copy of the page contained in the "test.html" file from the web server "www.server.de" located in Germany and displays that page in the web browser window instead of the original page which included the hyperlink.
Hyperlinks are like one way streets: while selecting a hyperlink in page A which links to page B enables a browser to display page B in place of page A, a user cannot link from page B to page A. This is because the hyperlink is merely a reference to the address of page B. It does not change page B in any way and page B's creator will therefore generally be unaware that page B has been linked to from page A.FN7
When a web browser fetches a copy of a page for display to the user, the browser also generally stores or "caches" a copy of the page on the computer system's hard disk.FN8 Caching pages increases the speed of fetching the same page again because the browser first checks to see if the page on the web server has changed; if it has not changed, the browser merely redisplays the cached copy of the page. The cached copies of the pages may also be browsed while the user is not connected to the Internet.
A graphical hyperlink may be created with the HTML markup:
<A HREF="http://www.server.no"><IMG SRC="http://www.server.no/logo.gif"></A>
This instructs a browser to fetch a copy of the logo image contained in the file "logo.gif" stored on the server in Norway and display it in place in the page which refers to it. This is sometimes referred to as "inline display" or "inline imaging" because a part of the virtual page, the image, is displayed as though it were part of the original page.
It should be noted that parts of a virtual page on a web server in Australia may reference parts of a page on a server anywhere else in the world provided only that the part in question has its own URL. However, the page is only assembled when a browser fetches a copy of it, plus any additional parts which it references, for display to the user.
Like a textual hyperlink, when the user selects the image, the browser fetches a copy of the default (home) page of the Norwegian server and displays it in the browser window instead of the original page which contained the hyperlink.
A simple frameset may be created with the HTML markup:
<frameset cols="12%,88%"> <frame name="frame1" src="http://www.test1.com.au/page1.html"> <frame name="frame2" src="http://www.test2.com.au/page2.html"> </frameset<
This instructs a web browser to divide the browser display window into two sections known as frames: "frame1" comprises the top 12% of the browser window, while "frame2" comprises the bottom 88% of the browser window. The browser fetches a copy of the file "page1.html" from the web server "test1.com.au" and displays it in frame1, then the browser fetches a copy of the file "page2.html" from the web server "test2.com.au" and displays it in frame2. The "page1.html" file could itself contain hyperlinks which, if selected, would display different pages in frame2 while frame1 remained the same.
What copyright is may be summed up as:
3.1 [...] a type of property right which is founded on a person's creative skill and labour. It is designed to prevent the unauthorised use by others of a work, that is, the original form in which an idea or information has been expressed by the creator. 3.2 Copyright is not a tangible thing. It is made up of a bundle of exclusive economic rights to do certain acts with an original work or other copyright subject-matter.FN9
The Australian Copyright Act 1968 (Cth) provides copyright protection for various subject matters which, for present purposes, include original literary and artistic works which were authored by Australian citizens and residentsFN10 and those persons whose countries afford Australian citizens and residents similar protection.FN11
A literary or artistic work is made when it is first reduced to writing or some other material formFN12 and is deemed to have been published if reproductions of the work have been supplied, whether by sale or otherwise, to the public.FN13 "Writing" is defined as a mode of representing or reproducing words, figures or symbols in a visible form.FN14 "Material form" is defined to include any form (whether visible or not) of storage from which the work or adaptation, or a substantial part of the work or adaptation, can be reproduced.FN15
In the case of a computer program, "adaptation" means a version of the work (whether or not in the language, code or notation in which the work was originally expressed) not being a reproduction of the work.FN16 In the case of other literary works, adaptation means: (a) a dramatic version of a non-dramatic work, or vice versa, whether or not a translation, (b) a translation of the work; and (c) a version of the work in which a story or action is conveyed solely or principally by means of pictures.FN17
"Literary work" is defined as including a table or compilation expressed in words, figures or symbols and includes a computer program.FN18 A "computer program" is itself defined as meaning an expression in any language, code or notation, of a set of instructions (whether with or without related information) intended to cause a device having digital information processing capabilities to perform a particular function.FN19
It is well settled that in the case of a literary work, copyright does not exist in ideas or information but in the order of the words, that is, the particular form in which the ideas or information are expressed.FN20 The originality required does not require that the expression be of original or inventive thought, nor that it show any literary or other skill or judgment, but that the work must originate from the author and not be copied from another work.FN21 It has also been accepted that where the expression of an idea is inseparable from its function, the expression forms part of the idea and is not entitled to copyright protection.FN22
Although it has been said that "what is worth copying is prima facie worth protecting",FN23 in the case of an insubstantial work literary copyright requires that the author has expended some skill, knowledge and creative labour in its creation. Thus, the titles of books and periodicals have been held to fail this test on the ground that the choice of the words involved too slight a degree of skill and labour to justify copyright protection.FN24
"Artistic work" means (a) a painting, sculpture, drawing, engraving or photograph, whether the work is of artistic quality or not; (b) a building or a model of a building, whether the building or model is of artistic quality or not; or (c) a work of artistic craftsmanship to which neither (a) nor (b) applies; but does not include a circuit layout.FN25
The Copyright ActFN26 grants the copyright owner of an original literary work the exclusive right to:
In the case of an original artistic work, the ActFN27 grants the copyright owner the exclusive right to:
In addition to the protection of the above "economic" rights, the Copyright Act also provides some limited protection for an author's "moral" rights.FN28 The Act creates a duty on a person in Australia not to attribute falsely the authorship of a work by inserting or affixing another person's name, initials or monogram in or on the copyright work, or a reproduction of the work, in such a way as to imply that the other person is the author of the work without the permission of the author.FN29 It should be noted that, as copyright may be assigned, the author of a work may not be the copyright owner.
Copyright in a literary or artistic work is infringed when a person who is not the copyright owner and who is not licensed by the copyright owner, does, or authorises the doing of, any of the acts comprised in the copyright.FN30
The onus of establishing the absence of a licence rests on the copyright owner.FN31 In this context a licence means no more than "permission" and may be an express or implied term of a contract or implied by conduct.FN32 For example, where copyright material is produced for a particular purpose, there is an implied licence to use that material to carry out that purpose.FN33 Also, it has been accepted that indifference may reach a stage where authorisation or permission may be inferred.FN34
There can be no authorisation of infringement unless there is an act of infringement of the kind allegedly authorised.FN35 However, a person does not authorise an infringement merely by knowing that another person might infringe the copyright and takes no action to prevent the infringement. Authorisation includes a mental element and it cannot be inferred that a person had, by mere inactivity, authorised something to be done if he neither knew nor had reason to suspect that the act might be done.FN36 Nonetheless, where a person expressly or impliedly extends an unqualified invitation to the public to exercise the exclusive rights of the copyright owner, it is unnecessary that the authorising person have knowledge that a particular act comprised in the copyright will be done.FN37
The Copyright Act provides that it is not an infringement of copyright in a literary or artistic work where there is:FN38 (a) fair dealing for the purpose of research or study; (b) fair dealing for the purpose of criticism; (c) fair dealing for the purpose of reporting news; (d) reproduction for the purpose of judicial proceedings or the giving of legal advice; or, in the case of a computer program, a reproduction for backup purposes. Additionally, there are a number of other statutory exceptionsFN39 to copyright infringement for various institutions including educational institutions and libraries in specified circumstances and for specified purposes.
It is clear that web pages can be the subject of copyright as original literary works which have been reduced to writing or some other material form. Additionally, web pages that have been marked up in HTML would arguably qualify as computer programs.
What is not so clear is whether a hyperlink which appears in a web page would itself qualify as an original literary work. Given the form of the HTML markup required, the name of the web server, the name of the document and the name of the directory in which it is stored, there is only one possible expression of the hyperlink for any particular page or image. Thus, it is arguable that, as a hyperlink expresses nothing more than information indicating the address or location of a particular web page or graphical image, it is not a literary work in which copyright can subsist. Alternatively, it is arguable that too little skill and labour have gone into its creation to justify copyright protection.
As a hyperlink is marked up in HTML and causes a device having digital processing capabilities to perform a particular function, it is arguable that the hyperlink is itself a computer program and therefore a literary work. However, it is submitted that the expression of a hyperlink is so bound up with its function, that it is not capable of being separated from its function and is therefore not entitled to copyright protection.
Where person A includes in web page A, a simple textual hyperlink to web page B authored by person B, there is no infringement by A of B's copyright in page B because A has not infringed the reproduction right, the publication right nor the adaptation right. The hyperlink in page A is only a reference to the address of page B which exists on B's web server.
Where another person, C, uses a web browser to view page A and selects the hyperlink in page A to view page B, C's browser fetches a copy of page B. This amounts to a reproduction of page B in the browser window as a copy of page B now exists in C's computer's memory if not also in the browser cache on C's hard disk. However, by making page B publicly available on a web server, the author of page B must be taken to have impliedly licensed third parties to use web browsers to view page B.
In the absence of an infringement of copyright by C, A cannot be found to have authorised C to make an infringing copy of B's page B by simply providing a hyperlink to it.
Nor is it arguable that the authorship of page B has been falsely attributed by A where the page containing the textual hyperlink bears no such annotation on which to found the implication of authorship.
Where person A includes in web page A, a graphical hyperlink to an image, the copyright in which is owned by person B, there is no infringement by A of B's copyright in the image because A has not infringed the reproduction right or the publication right. The hyperlink in page A is only a reference to the address of the image which exists on B's web server.
Where another person, C, uses a web browser to view page A, C's browser fetches a copy of the image while assembling page A and displays that image in place in page A. This amounts to a reproduction of the image in the browser window as a copy of the image now exists in C's computer's memory if not also in the browser cache on C's hard disk. However, by making the image publicly available on a web server, the owner of the image must be taken to have impliedly licensed third parties to use web browsers to view the image.
In the absence of an infringement of copyright by C, A cannot be found to have authorised C to make an infringing copy of B's image by simply providing a hyperlink to it.
However, it is arguable that where page A contains A's name, A may have falsely attributed the authorship of B's image because it appears to be part of A's page. Of course, A could avoid this by annotating B's image with the details of its true authorship if it was not already clear from the image.
Where person A includes in web page A, a hyperlinked frame which references web page B authored by person B, there is no infringement by A of B's copyright because A has not infringed the reproduction right, the publication right or the adaptation right. The hyperlink in a frame of page A is only a reference to the address of page B which exists on B's web server.
Where another person, C, uses a web browser to view page A, C's browser fetches a copy of page B and displays it in a frame in the browser window. This amounts to a reproduction of page B in the browser window as a copy of page B now exists in C's computer's memory if not also in the browser cache on C's hard disk. However, by making page B publicly available on a web server, B must be taken to have impliedly licensed third parties to use web browsers to view page B.
In the absence of an infringement of copyright by C, A cannot be found to have authorised C to make an infringing copy of B's page by simply providing a hyperlink to it.
However, it may be arguable that the display of B's page B in a frame of page A in C's web browser window falsely attributes the authorship of page B to A. This is because it may appear to C that the browser is displaying a page which is on A's web server and because page B is being displayed in the context of page A. Again, A could avoid this conclusion by making the authorship of page B clear in page A.
The technological means available to defeat rogue hyperlinks include: (1) requiring a password to gain access to pages; (2) encrypting pages so that a decryption mechanism is needed to view them; (3) using dynamic URLs which can be programmatically changed regularly; and (4) having the target web server validate the origin of hyperlinks.
The last method, in particular, is quite transparent and easily implemented in the web server software or as an extension to it. All web servers know the address of the page (the "referrer page") from which a person who has used a hyperlink has come. As the web server can verify whether the person used a hyperlink in one of the pages on its site or a foreign site, where the referrer page is not on the same or a related web server, the web server can either refuse to display the pageFN40 which is the target of the hyperlink or cause the web browser to fetch any specified page (eg the target web site's home page).
In the case of framesets, another option to prevent pages being "framed" by a third party is to include a script in the pages which removes any frames before it displays the pages.
The wide availability of the technological means to prevent anyone from creating a hyperlink to the publicly available pages and images of others on the web must surely reinforce the argument that those who make their pages and images available, without restriction, grant an implied licence to others to create hyperlinks to them.
It is submitted that the above discussion leads inevitably to the conclusion that the existing Copyright Act is quite able to cope with the phenomenon of the hyperlink in all its guises on the web. The current legislation does not overly constrain the continued development of the web, nor does it unduly prejudice the rights of copyright owners to control their web content. In cases where copyright owners wish to exercise greater control over their content than the legislation provides, the technological means are available to allow them to do so.
On 30 April 1998 the Government announcedFN41 a package of reforms to the copyright law which, among other things, are intended to improve copyright protection for material made available on the Internet. The existing technology- specific transmission rights are to be replaced with a new "broad-based technology-neutral right of communication to the public" which is subject to the existing raft of copyright exceptions. Full details of this new transmission right will not be known until the Government releases draft legislation. However, as the Government has stated that the proposed reforms implement many of the proposals contained in the discussion paper, Copyright Reform and the Digital Agenda,FN42 the following tentative observations may be made.
The discussion paper contains a proposal to exclude certain temporary copies of copyright material from the reproduction right where those copies are made in the course of a transmission to the public, unless the copies are used for any purpose other than simply being part of a technical process to enable viewing of the material, and that use constituted an exercise of copyright in the material requiring the permission of the copyright owner.FN43 This would clearly cover the copy created by a web browser in the computer's memory as it is used only to enable the technical process of viewing the page.
However, the discussion paper notes that "[d]ownloading a hard copy of the temporary copy or the transfer of such a copy to another file or onto a disk would be an exercise of the copyright owner's reproduction right and require permission" subject to the existing exceptions in the Act.FN44 Does this mean that the cached copies of web pages are excluded from the proposed exemption because they are transferred to a disk? Is the caching of web pages by a web browser simply part of a technical process to enable the faster or offline viewing of the pages? The answers to these questions will remain unclear until the draft legislation is released.
Footnotes:
FN1 There are estimated to be 29,670,000 individually connected computers in 240 countries. Wired News, February 16, 1997. Source: http://cyberatlas.com/market/usage/frequency.html.
FN2 The number of Internet users worldwide is estimated to be more than 100 million. "How Many People Online?", January 13, 1998. Source: http://www.cyberatlas.com/market/size/how_many_people.html.
FN3 "Information Management: A Proposal", March 1989, Tim Berners-Lee. See: http://www.w3.org/History/1989/proposal.html.
FN4 Netcraft's May 1998 Web Server Survey found 2,308,502 live WWW sites on the Internet. Source: http://netcraft.com/survey.html.
FN5 It is estimated that there are in excess of 320 million web pages online, excluding private, encrypted or password protected documents. Source: "Quantity or Quality?", Sorcha Ni hEilidhe, April 6, 1998, http://www.nua.ie/surveys/analysis/weekly_editorial/archives/1998/issue1no21.html.
FN6 The actual method of distinguishing the text as a hyperlink is left to the particular web browser and may often be a user configurable option.
FN7 Web server software may be configured to keep detailed logs which show the URL of the page from which a user who followed a hyperlink has come.
FN8 Web browser software may be configured to (1) check if a cached page has been changed each time it is requested, (2) to check only once per web browsing session, or (3) to never check if a cached page has changed.
FN9 Copyright Law in Australia - A Short Guide, October 1997, Attorney-General's Department. See: http://law.gov.au/publications/copyrightaus97.htm.
FN10 Copyright Act 1968, sec 32.
FN11 Ibid, sec 184.
FN12 Ibid, sec 22(1).
FN13 Ibid, sec 29(1)(a).
FN14 Ibid, sec 10(1).
FN15 Ibid.
FN16 Ibid, sec 10(1)(ba).
FN17 Ibid, sec 10(1)(a), (b), and (c).
FN18 Ibid.
FN19 Ibid.
FN20 Computer Edge Pty Ltd v. Apple Computer Inc (1986) 161 CLR 171 at p 181, per Gibbs CJ.
FN21 Victoria Park Racing and Recreation Grounds Co Ltd v. Taylor (1937) 58 CLR 479 at p 511, per Dixon J.
FN22 Autodesk Inc v. Dyason (1992) 173 CLR 330 at p 345.
FN23 University of London Press Ltd v. University Tutorial Press Ltd (1916) 2 Ch 601 at p 608.
FN24 Rose v. Information Services Ltd (1987) FSR 254 at p 255.
FN25 Copyright Act 1968, sec 10(1).
FN26 Sec 31(1)(a), 31(1)(c) and (d).
FN27 Sec 31(1)(b).
FN28 Sec 189-195A.
FN29 Sec 190, 193.
FN30 Copyright Act 1968, sec 36(1).
FN31 Avel Pty Ltd v. Multicoin Amusements Pty Ltd (1990) 171 CLR 88 at p 94-5.
FN32 Ibid.
FN33 Beck v. Montana Constructions Pty Ltd (1963) 80 WN (NSW) 1578 at 1580-1; Interstate Parcel Express Co Pty Ltd v. Time Life International (Nederlands) B.V. (1977) 138 CLR 534 at p 540, 547-8.
FN34 Performing Right Society v. Ciryl Theatrical Syndicate (1924) 1 KB 1, at p 9; cited with approval by McHugh J in Avel Pty Ltd v. Multicoin Amusements Pty Ltd (1990) 171 CLR 88 at p 123.
FN35 RCA Corporation v. John Fairfax and Sons Ltd (1981) 1 NSWLR 251, at p 256-8.
FN36 University of NSW v. Moorhouse (1975) 133 CLR 1, at p 12-3, per Gibbs J.
FN37 Ibid, at p 21, per Jacobs J.
FN38 Copyright Act 1968, sec 40-43A.
FN39 Ibid, sec 135A-135ZZH.
FN40 This method has been implemented by MSNBC's weather service to prevent hyperlinks in web pages originating at URLs other than http://www.intellicast.com from accessing the weather service's real-time weather radar maps as inline images. Wired News, "Intellicast Smartens Up to Banner Bypass", Kristi Coale, 28 March 1997, http://www.wired.com/news/topframe/2844.html.
FN41 "Copyright reform and the Information Economy", joint Media Release by the Attorney-General and the Minister for Communications, the Information Economy and the Arts, 30 April 1998, http://www.dca.gov.au/mediarel/98/064.html.
FN42 http://law.gov.au/publications/digital.htm, July 1997.
FN43 Ibid, para 4.55-4.61.
FN44 Ibid, para 4.60.