Blakeney, Michael; Macmillian, Fiona --- "Regulating Speech on the Internet" [1999] DigTechLawJl 6; (1999) 1(1) Digital Technology Law Journal 6

Regulating Speech on the Internet

Michael Blakeney and Fiona Macmillan

Asia Pacific Intellectual Property Law Institute
School of Law, Murdoch University, Western Australia

[e-mail]
[e-mail]

This paper considers the way in which traditional legal tools for regulating speech have been applied in the Internet environment. The paper commences by describing the aspects of this environment which are regarded as attracting regulatory interest. It then moves on to consider the way in which the regimes of defamation law, censorship law and intellectual property law impact on the Internet. Through this examination the paper raises the question of the appropriateness of the current approach of regulation by analogy.

1. Introduction

[1] The Internet has become one of the most powerful tools for disseminating information nationally, regionally and globally. Worldwide there are estimated to be over 50 million Internet users and this figure is expected to grow to 200 million by 1999. [1] This capacity for instantaneous communication, unconstrained by political or physical boundaries, has important implications for the way in which communications may be regulated by those public and private laws which impact upon free speech. These laws are typically concerned with the preservation of public order (blasphemy, obscenity, sedition and hate speech), protection of private reputation (defamation), and the preservation of proprietary interests in information (intellectual property). The Internet has not particularly affected the way in which those laws affect the content of communications, but it introduces new categories of defendants who may be pursued in cases of breaches of the substantive laws.

[2] Some appreciation of the organization of the Internet and the computer information facilities which it has spawned and the way that information is communicated on the Internet, is required to enable an appreciation of the categories of potential parties in actions seeking to regulate speech on the Internet.

2. The Internet

[3] The Internet is a world wide link of computers and computer networks which use a common protocol for communicating with one another. The physical connection between computers on the Internet is typically effected through leased space on existing telephone networks. The information content on the Internet is held in computers, known as `servers' which are owned and operated by information providers, which may be universities, government instrumentalities, or commercial enterprises. Computers which make up the Internet may be `host' or `client' computers. Host computers are connected to neighbouring Internet computers by a dedicated line through which messages are routed according to the Internet communication protocol. Client computers interface with host computers and provide the user with access to request and receive information from the Internet. Host computers are typically provided by commercial `Internet Service Providers (ISPs)'. Client computers may be directly connected to host computers through a local area network or may be connected to a host computer by a modem via the telephone network. A threshold issue to be considered in any examination of regulating speech on the Internet, is the respective liability for the content of Internet communications which may be imposed upon the operators of host computers, telecommunications carriers and ISPs.

2.1 The World Wide Web

[4] In 1990 the World Wide Web (WWW) system was devised for organizing and distributing documents on the Internet. This system enabled the storage of information without modification of its format and enabled the transmission of any type of data. A particular feature of the Web, which enhanced its utility, was the use of `HyperText Mark-up Language (HTML)' which enabled portions of documents to be linked together. The development of a `HyperText Transfer Protocol (HTTP)' enabled documents on different sites to be linked together. Two principal liability issues are raised by this facility. First, to what extent is the incorporation of material on another's site by hyperlinking a breach of copyright? Secondly, is a creator of a hypertext link in a document liable for the actionable content of documents with which the first document is linked?

[5] In 1993 the utility of the Web was considerably enhanced by the development of a `Mosaic', a browser software enabling easy access to Web information, utilizing the hypertext entries of Web documents. Mosaic was subsequently displaced by commercially available browsers such as `Netscape Navigator', which utilise the `look and feel' of the `Windows' software. An innovation with significant implications for regulating speech on the Internet was the development of the Platform for Internet Content Selection (PICS) by a World Wide Web Consortium (W3C) of non-governmental organizations, with offices in France, Japan and the USA. This body has formulated an industry standard which allows a Web site to exhibit labels concerning the content of Web material and enables PICS compatible software to respond to the labels. A consequence of this, which is discussed below, is that censorship will be achieved without the participation of the user, but imposed by commercial software houses.

[6] A final feature of the Web which has implications for the control of speech on the Internet is the increasing presence of sponsorship and advertising of Web sites. As with other commercial media, sponsors can both influence the content of communications and be held liable for material associated with them. A recent illustration of the latter, referred to by Edwards, [2] is the action brought by Jimmy Hill, the sports commentator, against a Scottish advertiser which sponsored a Web site which contained a forum where sports fans posted material defamatory of Mr Hill.

2.2 Domain Names

[7] Each host computer on the Internet is allocated a Uniform Resource Locator (URL) address which comprises four groups of numbers. To simplify this system in 1984 the Domain Name Space (DNS) system was developed. This system converts the numeric URL address into the hierarchical format of letters. The top level domains (TLDs) provide the general root systems under which the individual second level domain names (SLDs) are allocated to individual users. The principal TLDs are: `.com' (commercial enterprise); `.edu' or `.ac' (educational institution); `.org' (international organization); `.gov' (governmental organization. The allocation of domain names was originally undertaken in the USA, where of course the Internet originated. Until 1998, SLD allocation within these TLDs was operated by a private corporation, Network Solutions, Inc. With the entry of other countries into the Internet, system, SLDs have been allocated by national registries with a two letter country code designation. Thus http://www.murdoch.edu.au identifies the address on the Web of a host computer at Murdoch University, an educational institution in Australia.

[8] With the massive global expansion of the Internet, an Internet Ad Hoc Committee was established to report on the more efficient administration of the system. Its resultant report proposed the introduction of seven new generic TLDs (gTLDs); the introduction of a system of Registrars, and self-regulatory bodies to manage these TLDs; a trade marks policy and a dispute resolution mechanism. [3] These recommendations were incorporated into a gTLD Memorandum of Understanding (MoU) to be adopted by interested parties. The MoU proposed a global system of Registrars, with a Policy Oversight Committee and a Policy Advisory Body, formed from the signatories to the MoU. A particular problem which has been thrown up by the domain name system, is its overlap with trade marks. The effect of the allocation of a SLD, usually on a first come basis, [4] is to grant exclusivity to an individual wishing to establish a Web site utilising that SLD. This has obvious implications for the freedom of commercial speech on the Internet and has raised questions of liability on the part both of the SLD applicant and the granting authority. The granting authorities in each country are designated as such by the Internet Assigned Numbers Authority (IANA), which had been set up by the Internet Society, a non-profit body established in the USA to coordinate the operation of the Internet. [5]

2.3 Electronic Mail

[9] The domain name system enabled the establishment of three significant Internet facilities with important implications for liability for the content of communications on the Internet: electronic mail, Web sites and bulletin boards. Electronic mail (e-mail) permits communications on the Internet to anyone at a given domain address identified by a `userid' preceding the domain address. For example, [e-mail] is the e-mail address of the current Vice Chancellor of Murdoch University. E-mail has become one of the most powerful communication facilities on the Internet. Users of e-mail have tended to be less disciplined in formulating the content of their messages than those utilizing other communication methods. A consequence of this has been the greater propensity to disseminate actionable material, particularly when a message is forwarded through the Internet. A user wishing to communicate with a large number of people at the same time can utilize a `List Server' which will automatically send a single message to every person on a mailing list, such as every member of a university community. Where an actionable communication is launched through a List Server, an issue of the liability of the provider of this facility is raised.

2.4 Bulletin Boards and Usenet Newsgroups

[10] The second facility which the Internet makes available is the Web site, which provides a location at which information about the maintainer of the site can be made available. Web sites have typically been established by government departments, international organizations, educational institutions and commercial enterprises. Liability may arise from communications which are placed on a Web site. These communications may be via bulletin boards or Usenet newsgroups.

[11] Computerised bulletin boards, known as BBS, are a facility established on a host computer by which messages can be placed to be read by people accessing the BBS. These systems may be operated by commercial enterprises, requiring payment for access, others are freely available to all users. Bulletin board systems may also be used for the sending of electronic mail and for on-line discussions. A BBS may also make computer files available to users, for example various software applications may be distributed to interested users. Both providers and users of BBS may be liable for the content of messages placed on bulletin boards.

[12] Newsgroups are discussion lists made up of comments which are provided by subscribers and sorted according to subject matter. Collectively these Internet newsgroups are known as `Usenet'. In September 1996 there were estimated to be some 14,000 Usenet newsgroups [6] posting millions of articles instantaneously to users around the world. The liability implications of this volume of correspondence is self evident.

3. Defamation

[13] The anarchy and libertarianism which characterised the early life of the Internet, when users were either at universities or in the computer industry, produced an ethos in which defamation flourished. The practice of `flaming', or abusing people through these communications was apparently an accepted form of discourse, for which the expected sanction was retaliation in kind. With the expansion of Internet use, community standards for speech began to intrude. However, the first Australian Internet defamation case, Rindos v Hardwick [7] concerned material placed on a bulletin board maintained by an anthropological science association which was in the old flaming style. The plaintiff was a US-born anthropologist who had lectured in that subject at the University of Western Australia until June 1993. In March 1993 the University's Tenure Review Committee recommended that the plaintiff be denied tenure on the ground of insufficient productivity and he was dismissed by the Vice Chancellor of the University with effect from 13 June 1993. On 23 June 1993 a message written by Hugh Jarvis, a US- based anthropologist, appeared on the DIALx science anthropology bulletin board criticising the University of Western Australia for dismissing the plaintiff. This bulletin board was available to subscribers around the world and had attracted some 23,000 subscriptions. Most universities subscribed to the bulletin board and it had a wide international readership. On 26 June 1993, in reply to the message published by Hugh Jarvis, the defendant transmitted a message to the bulletin board from Derby on the north coast of Western Australia. This message criticized, inter alia, the plaintiff's research methodology in relation to the anthropology of Aboriginal culture and referred to rumours concerning the plaintiff's sexual practices. The trial judge accepted that these two matters were defamatory of the plaintiff, that they caused serious harm to the plaintiff's personal and professional reputation and that they caused him a good deal of personal hurt. He awarded the sum of $40,000 in damages.

[14] Because both the plaintiff and the defendant were both within the Western Australia no question was raised as to which defamation law was applicable to these facts. It is possible to conceive of a situation where material placed on the Internet might be defamatory in the jurisdiction of the plaintiff, but defensible in the jurisdiction of the defendant. Another issue which was not canvassed was the liability of the bulletin board proprietor, the ISP and the telecommunications carrier. This issue was considered in the UK in an action brought by Dr Laurence Godfrey, a lecturer in physics and computers science who has sued Demon Systems, a British ISP in relation to allegedly defamatory statements posted by a student to a usenet news group. [8] Where the liability of a carrier or an ISP in defamation requires knowing communication, there is a good argument that given the vast volume of messages passing through their hands, it would be unreasonable to impute knowledge of the content of those communications. [9]

[15] This issue was addressed in the US decision Cubby, Inc v. CompuServe, Inc. [10] , which concerned defamatory material distributed by CompuServe, which operated an electronic library to its subscribers. Subscribers could obtain access to over 150 special interest forums. One of these was a journalism forum, which contained material defamatory of the plaintiff. The court held that as the system operator had no knowledge of the defamatory material and exerted no editorial control over it, there could be no question of liability. The court held that the liability of an electronic news distributor such as CompuServe, should be equated to that of a public library, bookstore or news stand. A significant feature of the case was the imposition of the onus on the defendant to establish that it had no reason to know of the defamatory statements.

[16] CompuServe should be contrasted with Stratton Oakmount, Inc and Porush v Prodigy Services Company [11] which concerned a computer network operated by Prodigy, with at least two million subscribers. One of its facilities was `Money Talk', a bulletin board to which subscribers could post statements regarding financial matters. Prodigy held itself out as exercising editorial control over the content of messages posted to the bulletin board and it promoted its service as `family oriented'. The plaintiff succeeded in an action for defamation against Prodigy in relation to matter posted to the bulletin board. In response to Prodigy's claim that it was not possible to take editorial responsibility for the 60,000 messages per day which were posted to its bulletin board, the court observed that Prodigy had held itself out as so doing and it implemented editorial controls through screening software. In distinguishing CompuServe, the court ruled that Prodigy exercised sufficient editorial control over its bulletin boards to render it a publisher with the same responsibilities as a newspaper. An obvious implication of this decision is that to avoid liability a computer service provider should exercise as little editorial control as possible. An issue which was raised indirectly was the ability of a remotely located defendant to defend a defamation action. In Rindos v Hardwick the court permitted substituted service. The defendant wrote to the court explaining that he had insufficient resources to travel to Perth to defend the action. This problem would be compounded for defendants outside the jurisdiction in matters arising from defamation on the Internet. However it would probably be balanced by the practical difficulty of enforcement against an extraterritorial defendant.

[17] Given the immediacy of speech on the Internet, particularly that involving `chat lines' on interactive bulletin boards, a question has arisen as to whether defamatory statements in this context is libel or slander. In relation to libel, a plaintiff does not have to prove damage. The law will assume that some damage will flow from the act of publication in a permanent form. On the other hand, slander requires proof of actual damage. In the broadcasting context in Australia, by way of example, radio and television transmissions are deemed to be `publication in permanent form' exonerating a defendant from the necessity to prove actual loss. [12] Similar legislation may be required to clarify the situation in relation to speech on the Internet.

[18] In the Australian federation there are defamation laws in each State and Territory. As these are not uniform, plaintiffs will shop for the forum which is most advantageous for them. For example, in New South Wales, a defendant seeking to rely on the defence of justification, must also prove that the defamation was in the public interest. The onerousness of this defence, compared with those jurisdictions where the public interest requirement is not imposed, provides an advantage for plaintiffs. Where defamation comprises speech on the Internet, which is transmitted to each State, a plaintiff will be able to select the favourable jurisdiction. This will also apply to international transmissions on the Internet where plaintiffs will be expected to forum shop. The common law position on appropriate forum requires an examination of the forum in which the action has the most real and substantial connection. [13] The choice was superficially straight forward in the Rindos case, where both parties were present in Western Australia. However had the plaintiff academic returned to his US home, or to the UK, where he had most recently worked, the choice of forum might not have been as simple. [14]

[19] The immediacy of e-mail communications has precipitated a style of speech which has caused actionability in a number of different areas of law. The circulation of tasteless jokes and rumours by employees within an enterprise has caused problems in the areas of sexual harassment, as well as defamation. However, it has not only been employees which have generated actionable speech. A recent settlement in the UK concerned an action brought by Western Provident Association against Norwich Union Healthcare for spreading untrue rumours on its internal e-mail system about Western's financial stability. [15]

4. Censorship of Obscene and `Unsuitable Material'

4.1 Federal Initiatives

[20] As a federation, Australia has both Federal and State parliaments. In a collision between State and Federal laws, the latter will prevail. The matters on which the national Parliament can legislate are listed in the Constitution. The Constitution, which dates back to 1901, when the Commonwealth of Australia was established, lists the legislative subjects which the Founding Fathers, considered appropriate matters for national action. Matters outside the Federal Constitution are generally subjects for State legislation. Obviously, the computer and the Internet was not within the contemplation of the Founding Fathers. Federal governments have taken the position that the Internet is equivalent to telecommunications and broadcasting, over which it has unquestioned legislative authority. However, the States, taking the position that Internet communications is not a subject within the purview of the Constitution, have introduced their own legislation regulating the content of Internet communications.

[21] The precipitant of parliamentary interest in censorship of Internet communications has been expressions of public concern about the availability of unsuitable material to minors. In 1991 a Senate Select Committee on Community Standards Relevant to the Supply of Services Utilising Electronic Technologies had been established as a result of expressions of community concern about the content of some recorded telephone services. [16] In October 1993 the Select Committee issued a report on video and computer games in which it warned of the availability of pornographic and violent material on bulletin boards. [17] In November 1993, the Federal minister for Communication and the Arts established a Task force on the Regulation of Computer Bulletin Board System to examine regulatory options. In September 1995 the Select Committee, following a public seminar and public consultations, issued Part 1 of its report on the regulation of computer on-line services. This report reviewed some of the difficulties in regulating on-line content, particularly that emanating from outside the country. In November 1995, following consultation with technical consultants, the Select committee issued Part 2 of its report. Eight recommendations were made, including: (i) the introduction of legislation to criminalise the use of a computer service to transmit, obtain possession of, demonstrate, advertise or request the transmission of restricted material; (ii) access and service providers be required to verify the identity of all clients and that all clients are over 18; (iii) that an industry self-regulation code be established to deal with on-line content.

[22] The activism of the Senate Select Committee probably reflects the influence in that House of the conservative moral influence of an independent Senator holding the balance of power between the two major political groupings. An attempt by the Federal government to seize the initiative in this area was a direction in June 1995 by the Minister for Communications and the Arts to the Australian Broadcasting Authority (ABA) to investigate the content of on-line services, including services on the Internet. In December 1995 the ABA released an Issues Paper calling for submissions from interested parties. [18] Following public consultations on 30 June 1996, the ABA reported to the Minister recommending the development of industry codes of practice by Australian ISPs and the registration of those codes with the ABA and the establishment of an Online Labelling Task Force to consider the development of Internet content labelling schemes, compatible with the Platform for Internet Content Selection (PICS).

[23] On 28 August 1997 the Minister issued a direction to the ABA to investigate and advise on matters to be included in industry codes of practice and to advise on national and international developments in the use of on-line content labelling services. [19] Among the organizations which are developing systems to interact with PICS is the Recreational Advisory Council for the Internet (RASCi). The RASCi labelling system grades the level of violence, sex, nudity and language on a rating of 0 to 4. A user's computer can be set to block out material which exceeds a prescribed rating. This labelling system has been incorporated into a number of commercial Web browsers, such as Microsoft's `Internet Explorer'. [20] In September 1996, Demon Internet, the largest UK ISP, announced that it would require all users to rate their Web pages using the RASCi rating system. [21]

[24] A particular development noted with approval by the ABA is the growth or e-mail hotline services set up by industry associations and children's interest groups to receive and investigate sites alleged to contain child pornography. The activities of the UK Internet Watch Foundation (IWF) in this regard, has been noted with approval by the Manager of the On-line Services of the ABA. [22] The ABA has called for the establishment of a similar alliance between industry and the police in Australia.

[25] A separate actor at the Federal Parliamentary level has been the Senate Select Committee on Community Standards Relevant to the Supply of Services Utilising Electronic Technology. This Committee, espousing deeply held concerns about community moral standards, issued a report in June 1997 recommending fines of up to $100,000 per offence for the electronic transfer of pornographic or other `offensive material'; that on-line service providers be held responsible for on-line material; that the States and territories criminalise the transmission of objectionable material through computer on- line services; and that specially designated State and Territory police units conduct random on-line checks to detect illegal activity. [23]

4.2 State and Territory Legislation

[26] Responding to community concerns with material carried on the Internet which might be accessed by minors, State and Territory legislatures have introduced, or are contemplating the introduction of criminal laws dealing with on-line content. Some attempt has been made to coordinate this legislative activity through the Standing Committee of Attorneys General (SCAG). In 1996 the New South Wales Parliamentary Counsel's Office drafted Model Offence Provisions for the consideration of SCAG. These provisions propose that it be an offence to send, receive, permit access to or retrieve `objectionable material' through an `on-line service'. The imposition of criminal liability upon ISPs attracted considerable industry criticism [24] and SCAG has delayed in coordinated action. In the meantime Victoria, Western Australia and the Northern Territory have passed legislation directed at on- line content. The Western Australian and Northern Territory legislation creates offences of transmitting `objectionable material' or `restricted material' to a minor. `Objectionable material' embraces matter which would typically be classified as pornography. [25] `Restricted Material' is defined in the Western Australian Act, for example, as including `an article that a reasonable adult, by reason of the nature of the article, or the nature and extent of references in the article to matters of sex, drug misuse or addiction, crime, cruelty, violence or revolting or abhorrent phenomena, would regard as unsuitable for a minor to see, read or hear' [26] . Article is defined as including a film and `film' is defined as including a `computer generated image', produced by the use of a computer on a computer monitor. Any legislation which applies a reasonable adult test is going to be problematic for free speech. This was illustrated in the parliamentary debates on these statutes. The opposition parties tended to take the classical libertarian position, thus the principal opposition speaker in the Western Australian Legislative Assembly noted that `all of us who are adults think we are reasonable, even if we have grave doubts about the reasonableness of others'. [27] She expressed concern with the State intervening, under a vaguely expressed statute, in a matter which more properly lay within the province of parental control. However on both sides of Parliament, the subject of child pornography on the Internet was identified as the justification for this legislation. Some concern was expressed with the imposition of strict liability. The Victorian legislation, however, imposes upon the prosecution the onus of proving that on-line objectionable material or material unsuitable for minors was `knowingly' created, published, transmitted or made available.

[27] Commentators have called into question the constitutionality of both the State and Territory legislation, on the one hand, as well as the Federal legislation on the other. [28] In its 1991 decision in Nationwide News Pty Ltd v Wills and Australian Capital Television v The Commonwealth [29] the High Court of Australia had identified an implied constitutional freedom of political discussion. This freedom was affirmed by the High Court in its 1997 decision in Lange v Australian Broadcasting Corporation. [30] The State and Federal legislation in proscribing the on-line dissemination of `objectionable material' may in its application to politically objectionable material be unconstitutional. The recognition of the public interest in freedom of political discussion has been recognised, for example, in the intellectual property context. [31] Thus in Davis v Commonwealth of Australia [32] the attempt by the Australian Bicentennial Authority to restrict the use of the term `200 years' in relation to use of the slogan `200 Years of Suppression and Depression' by Aboriginal applicants, was disallowed by the High Court as a legitimate `pungent protest'. [33]

[28] The application of US constitutional authorities to Australian circumstances is problematic, since in the US the various freedoms which it guarantees are expressly contained within its various amendments. However, the persuasiveness of US precedent has been noted in a number of cases, including the High Court's consideration of the implied right of political speech. It will be interesting to note the influence of the Supreme Court's recent consideration in Reno v American Civil Liberties Union [34] of the application of constitutional freedoms to speech on the Internet. That case concerned the constitutionality of two provisions of the Communications Decency Act of 1996 (CDA) which sought to protect minors from `indecent' and `patently offensive' communications on the Internet. Section 223(a)(1)(B) of the CDA created a criminal offence for anyone who, by means of a telecommunications device, knowingly makes, creates or solicits, and initiates the transmission of any `comment, request, suggestion, proposal, image, or other communication which is obscene or indecent, knowing that the recipient of the communication is under 18 years of age, regardless of whether the maker of the communication placed the call or initiated the communication.' Section 223(d) made it a criminal offence for anyone who uses an interactive computer to send to a person under 18, or to display in a manner available to a person under 18, any material that is patently offensive as measured by `contemporary community standards'. The Supreme Court ruled against the legislation on a number of grounds which may well be applicable to similar legislation in other jurisdictions. The court found that the CDA placed an unacceptable burden on speech, as there were less restrictive means, such as through the use of appropriate software that the protective object of the statute could be achieved. Also the Court observed that it was not technologically, nor economically feasible for ISPs to screen the ages of recipients of Internet communications. In other words the Court ruled that there were more effective and less restrictive means of screening speech on the Internet. The Supreme Court endorsed the decision of the trial judge who had observed that `if the goal of our First Amendment jurisprudence is the `individual dignity and choice' that arises from `putting the decision as to what views shall be voiced largely into the hands of each of us', then we should be specially vigilant in preventing content-based regulation of a medium that every minute allows citizens actually to make those decisions'. [35] As a policy matter the Supreme Court stated,

The record demonstrates that the growth of the Internet has been and continues to be phenomenal. As a matter of constitutional tradition, in the absence of evidence to the contrary, we presume that the government regulation of the content of speech is more likely to interfere with the free exchange of ideas than to encourage it. The interest in encouraging freedom of expression in a democratic society outweighs any theoretical but unproven benefit of censorship. [36]

[29] To deal with the strong expressions of public concern which followed the Supreme Court decision in Reno v ACLU, [37] President Clinton on 16 July 1997 convened a White House summit on Internet Parental Empowerment Tools. [38] This summit discussed the various filtering tools available to concerned parents. Although the use of these tools has been represented as a way of removing Internet access from regulatory interference, [39] the American Civil Liberties Union expressed concern that the institution of ratings systems as a default in browser software would have the effect of imposing the censorship views of software companies. [40] Additionally, the ACLU pointed to the practical difficulty of rating conversation on the Internet. Because at least half of the Internet traffic emanates from outside the USA, the ACLU expressed concern that this speech would be denied to local users. [41]

5. Intellectual Property Controls

[30] Intellectual property laws have been characterised as the conferral of proprietary rights in information. [42] This conferral has been achieved through national legislation enforceable in national courts. The digitisation of reprographic and communications technologies, together with the development of the Internet has obviously had a profound impact upon intellectual property law. These technologies have created new possibilities for the wrongful appropriation of intellectual property rights, while rendering it difficult both to detect infringements and to bring infringers within national jurisdictions. The principal fields of intellectual property law which have significantly impacted upon the freedom of speech on the Internet are copyright and trade marks law.

5.1 Copyright

[31] The conversion of copyrighted works into digital form and their dissemination over the Internet raises obvious copyright issues, but the Internet technology itself creates new forms of protectable works and new possibilities for infringement. In addition to the existing categories of works in which copyright has traditionally subsisted, copyright may subsist in the structure and layout of a Web site or Web page, including its `look and feel', in the software generating fonts [43] and possibly also in respect of hypertext links with other Web sites or Internet materials. [44] A hyperlink is typically an underlined expression which when activated by the user's mouse retrieves and displays the file identified by the underlined expression. The recent Scottish case, The Shetland Times v Wills [45] , which imposed copyright liability upon the owner of a Web site which incorporated references to the pursuer's headlines, is cited as an example of copyright liability which may arise from hypertext links. The copyright infringement in that case arose from the copying of the pursuers headlines, which were copyright works in themselves. It has been suggested that in placing materials on the Internet there is an `implied public access' to that material, even through hypertext links. [46] On the other hand, it has been suggested that the extent of the licence to use source materials on the Internet should be spelt out on the owner's Web page. [47]

[32] The construction of a Web site may involve the inadvertent copying of another's works. For example in Marobie v NAFED [48] an US association of fire equipment distributors was found liable for copyright infringement in ornamenting its Web site with clip art which was owned by a software company.

[33] An issue with great significance for the regulation of speech on the Internet and which was unresolved at the WIPO diplomatic conference in December 1996, was whether the temporary storage of material into a user's computer represented an infringement of copyright. [49] The critical issue is whether the downloaded file could be considered a material fixation, or copy of the original work. The US Copyright Act, Title 17 of the US Code, in s.101, defines a copy as a fixation `from which the work can be perceived, reproduced or otherwise communicated, either directly or with the aid of a machine or device'. To meet the objection that this might stifle access to speech on the Internet is the fair use defence which is available for educational and critical uses of Internet material. Among the factors which will be considered in deciding the application of the fair use defence will be the substantiality of the copying in relation to the work as a whole and the purpose and character of the use. The courts have generally condemned the unauthorised distribution through the Internet of works available for commercial sale. Thus in Playboy Enterprises Inc v Frena [50] the publishers of Playboy magazine, in enjoining the distibution of scanned images from the magazine on a bulletin board, were able to point to the commercial impact of that unuathorised use both on sales of the magazine and upon their own plans to distribute the magazine on the Internet. This case also emphasizes the fact that the fair use defence is available to users of Internet material and not necessarily for the provider of the archive of the accessed material. [51]

5.1.1 Authorising Communications to the Public

[34] Original literary, dramatic, musical or artistic works, sound recordings, films, broadcasts and cable programmes which are placed onto the Internet without the authorisation of the owners of those works will be a breach of copyright. This result did not require any modification of the principles of intellectual property law, however it has been placed beyond doubt by the promulgation in the 1996 WIPO Copyright Treaty [52] which provides in Article 8 in relation to artistic and literary works, that authors `shall enjoy the exclusive right of authorising any communication to the public of their works, by wire or wireless means, including the making available to the public of their works in such a way that members of the public may access these works from a place and at a time individually chosen by them.' The current Australian proposals for implementing Article 8 of the WIPO Copyright Treaty, [53] are contained in the Discussion Paper , Copyright Reform and the Digital Agenda. [54] This Paper proposes the replacement of the Copyright Act's existing wireless broadcasting and diffusion rights with a `transmission right' and a `right of making available to the public'. The transmission right is intended to be a `broadly-based technology-neutral . right' [55] , which would apply to transmissions to the public in the traditional non-interactive sense of `broadcasting', that is the emitting of signals from a transmitter to a receiving device at a time chosen by the person making the transmission. The person receiving a broadcast can only receive it at a time when the person making the broadcast chooses to make the transmission. [56] The right of `making available to the public' is described as being exercised `when copyright material was made available to the public in such a way that it could be accessed at a time and a place chosen by members of the public. This right is designed to cover interactive on-demand services.' [57]

5.1.2 Liability of Communications Carriers

[35] The issue of the liability of communications carriers for secondary breaches of copyright arising from the carriage of protected works was recently considered by the Australian High Court in Telstra Corporation Ltd v Australasian Performing Right Association Ltd. [58] This case concerned music played to users of the telephone system while they were on hold. The Australian Performing Rights Association contended that playing music on hold to persons using mobile phones constituted a breach of the exclusive right of the copyright holder in s 31(1)(a)(iv) of the Copyright Act 1968 (Cth) to broadcast a literary, dramatic or musical work and that playing music on hold to persons using conventional phones breached the exclusive right in s 31(1)(a)(v) to cause the work to be transmitted to subscribers to a diffusion service. Both of these arguments were sustained by a majority of the High Court, although the second one proved to be the more contentious of the two. [59] The Discussion Paper suggests that telecommunications carriers ought not be responsible either as infringers of the copyright of material which may be accessed by the public [60] , or as authorisers of infringements. [61]

[36] The principal Australian authority in this area is University of NSW v Moorhouse [62] which held the appellant liable as authorisers for providing photocopy machines for general use by patrons of the library, without adequate notices warning against use of the machines to infringe copyright. The court held that `authorise' in this context means `sanction, approve or countenance'. Although that case did not require actual knowledge of the infringements taking place using the apparatus supplied by the authoriser for liability to arise, it does seem that some ability to control infringement is required. If, on the basis of the mere supply of a photocopying machine without adequate monitoring or warnings, a library can become liable as for authorising infringement, it is quite unclear why the person who provides the infrastructure for a transmission, including generating the relevant electro-magnetic impulses or other signals, would be in the clear. [63] This possibility ought to be taken seriously by legislators because if it is possible, the desirability of communications providers as defendants means that some plaintiff (probably a collecting society) will try to make it a reality.

5.1.3 Liability of Internet Service Providers (ISPs)

[37] The position of ISPs for authorising breaches of copyright has been described as `conceptually distinguishable' from the position of communications carriers. [64] The reason for this is that ISPs do have a considerably greater ability to monitor material on their systems and to remove infringing material.

ISPs are the gatekeepers of the Internet. They come in all sizes . from well- known continental wide networks such as Netscape, Netcom, America Online and Compuserve, to hundreds of small and medium-sized ISPs that serve local and regional constituencies. All have one thing in common - a customer base that needs them to get onto the information highway.

. They have the ability, of course, to dictate the online environment in which their customers are operating through the use of written control policies and guidelines. Furthermore they can utilise and implement technology, including software, that is capable of automatically screening material on the network. [65]

[38] From the point of view of the aims of copyright law, there would be some concerns if the result is to mandate the complete removal of certain copyright material. Also, if ISPs are required closely to monitor material then this may have a rather obvious negative effect on privacy and freedom of speech on the Internet. Such a negative effect might not, of course occur, if the liability on ISPs is imposed not with the rather blunt and old-fashioned instrument of liability for authorisation, but with a more sophisticated instrument, such as a qualified obligation to remove material on notice from copyright holders. [66] It is fair to say that the jury is still out on the question of whether liability should be imposed on ISPs and what form that liability might take. [67]

[39] Certainly the position is clearer in relation to infringements of which ISPs are aware. In the US, in this situation an ISP would be liable as a contributory infringer. [68] It has been suggested that under UK law an ISP might be in the position of a joint infringer as it has the opportunity to monitor the material which it carries. [69]

5.2 Trade Marks

[40] A number of disputes have arisen where companies with similar names, or manufacturing the same sorts of products have sought to adopt the same or similar domain names on the Internet. With the introduction in most countries of the protection of well-known marks, as an obligation enjoined by the Agreement on Trade Related Intellectual Property Rights (TRIPs), [70] reputed traders have successfully prevented the use of domain names by unauthorised third parties which are identical or substantially similar to well known marks. [71] A particular problem for trade marks protection is that trade marks registration can be obtained in respect of each of 42 classes of goods and services, whereas only a single domain name is available for commercial enterprises.

[41] An illustration of the sorts of problems which can be caused by the application of trade marks principles to the acquisition of domain names is provided in Avnet Inc v Isoact Ltd. [72] The plaintiff was a distributor of electronic components and computer software. It maintained a web site and published trade catalogues. It conducted its business by reference to its `Avnet' trade mark and had obtained a registration for this mark in the UK in class 35. This class covers `advertising and promotional services'. The defendant conducted an entirely different business as `Aviators Network', an ISP with a particular focus on aviation matters. The defendants used the domain name `avnet.co.uk'. The plaintiffs, alleging an infringement of its class 35 trade mark, sought an order that the defendant assign to it the defendant's domain name.

[42] The trial judge, Jacobs J, refused to make the order sought by the plaintiff as he did not consider the defendant to be engaged in any activity which infringed the plaintiff's trade mark. He ruled that although persons might use the defendant's web site for advertising purposes, this was not the essence of what the defendant did, but was incidental to its provision of a forum for the discussion of aviation matters.

[43] Jacobs J had expressed doubts as to whether any Internet users on accessing the defendants aviation web site, would have been confused into thinking that there was some association with an electronics components distributor. [73] Contrasting domain names with trade marks, Jacobs J., noted that the former operated on words alone and not words connected with goods and services. Consequently, users of the Internet know that when a word is searched `even if a searcher is looking for the word in one context, he will, or may find web pages or data in a wholly different context'. [74] Had the plaintiff succeeded in its application, a consequence would have been to use a UK-based proprietary right as a means of obtaining a domain name, conferring global exclusivity.

[44] A number of domain name disputes have arisen in the context of what has been called `Cybersquatting', where individuals obtain domain name registrations which are likely to be sought by well known enterprises. [75] For example, in Harrods Limited v Network Services Limited, [76] the well known department store sought an injunction to prevent the UK domain name allocation authority from allocating the domain name `harrods.com.' to an unrelated third party. The trial judge accepted that passing off and trade marks principles were applicable to domain names and ordered the unauthorised third party to relinquish the domain name.

[45] The Memorandum of Understanding on generic TLDs, which was promulgated in February 1997 proposes that where a domain name is sought for `an alphanumeric string that.is deemed to be internationally known and for which demonstrable intellectual property rights exist, may be held or used only with the authorisation of the owner'. As a matter of practice this will apply to trade marks registered in at least 35 countries in at least four of the seven geographical regions of the world, comprising: North America, Latin America, Western Europe, Central and Eastern Europe, the Baltic States and CIS, Africa, the Middle East and Asia. [77] To adjudicate disputes concerning the allocation of domain names, the World Intellectual Property Organization (WIPO) has established Administrative Domain Name Challenge Panels (ACPs). [78] A problem which this structure shares with the administration of the Internet generally, is that it is consensual, it exists as a consequence of the MoU and does not preclude resort to national courts.

[46] In any event, the resolution of disputes concerning the warehousing of well-known marks is much simpler in practice than the resolution of disputes between less well-known, bona fide traders in related areas of business.

6. Categorising Speech on the Internet

[47] The imposition of liability for actionable speech which is accessible on the Internet depends in large part on the way in which the Internet is classified as a communications medium. Analogies abound, constrained only by the imagination of the commentator. President Clinton referred to the Internet as `our new townsquare'. [79] Similarly, in Reno v ACLU [80] the Supreme Court commented that `through the use of chat rooms, any person with a phone line can become a town crier with a voice that resonates farther than it could from any soap box. Through the use of web pages, mail exploders, and newsgroups, the same individual can become a pamphleteer.' [81]

[48] As the technological possibilities of the Internet have evolved, so the analogies have broadened. When the Internet was confined to static text and pictures, the most apposite analogies were those associated with publishing and the print media. Thus, the Internet defamation and pornography cases focussed on the knowledge, which an editor or publisher might reasonably be expected to have. When it became apparent that the vast amounts of data carried on the Internet made this knowledge unrealistic, the analogy shifted to that of a librarian, who would by and large be ignorant of the content of publications held in his collection. The creation of search engines for the retrieval of information reinforced this analogy.

[49] With the introduction of e-mail services, the imagery of the postal service began to be employed. In this context, the imposition of liability upon a postal service for the content of messages carried by it was considered inappropriate. With the introduction of bulletin board systems, initially, the position taken by the courts was the same as in relation to traditional bill boards, where the proprietor of a property on which a bill board was located would be held liable only for actionable messages brought to the proprietor's attention. [82] The development of interactive bill board systems generated the image of the Internet as a telecommunications system or as a broadcasting system. In Australia, it will be recalled that Federal government attention grew out of a concern with pornographic telephone chat services. The broadcasting imagery was thought appropriate because of well developed principles of the public responsibilities of broadcasters, particularly to minors. However, as the responsibilities of public broadcasters originally arose from the perception of fiduciary obligations generated by the limited availability of the broadcasting spectrum, this analogy is probably strained in its application to the Internet, which has no equivalent limits. It is probably not too facile to observe that the computer monitor and the capacity of the Internet to carry audiovisual images has generated the analogy with television. Because of the increasing use of encryption on the Internet, cable television is the best analogy here.

[50] Because the Internet is a combination of communications media, there is a strong case for arguing that the result is greater than the sum of its parts and that the concept of liability should be viewed from a sui generis perspective. The dynamic evolution of the Internet suggests that it will be carrying forms of speech very remote from the traditional regulatory categories. The development of rating and blocking technologies will probably shift responsibility for the control of speech on the Internet into the home and away from governments.

---

1 Reno v American Civil Liberties Union, Supreme Court (USA), Case No.96-511, 26 June 1997, [Internet] URL: http://www.bna.com/e-law/cases/reno0627.html (at 10). Back to Text.

2 Edwards, `Defamation and the Internet: Name Calling in Cyberspace', [Internet] URL: http://www.law.ed.ac.uk/c10_main.htm. Back to Text.

3 International Ad Hoc Committee, Recommendations for Administration and Management of gTLDs, [Internet] URL: http://www.gtld-mou.org/draft-iahc-recommend-00.html. Back to Text.

4 See Stoodley, `Internet Domain Names and Trade Marks', [1997] 9 EIPR 509. Back to Text.

5 See [Internet] URL: http://www.isoc.org:what-is-isoc.html. Back to Text.

6 Edwards, n.2 supra. Back to Text.

7 Supreme Court of Western Australia, Unreported, 31 March 1994. Back to Text.

8 Loundy, IBSA Intellectual Property Section Council Internet Law Subcommittee Report, 21 June 1998, [Internet] URL: http://www.neutral.com. Back to Text.

9 See Charles, `Computer Bulletin Boards and Defamation: Who Should be Liable?', (1987) 2 Jnl Law & Technology 121. Back to Text.

10 776 F. Supp 135 (1991). Back to Text.

11 Supreme Court, State of New York, Oct. 1995, noted in Platford, `Loose Words on the Net: Defamation and Contempt in Cyberspace', [Internet} URL: http://www.gtlaw.com.au/gt/pubs/loosewords.html. Back to Text.

12 Broadcasting and Television Act, 1942 (Cth), s.120. Back to Text.

13 The Abidin Daver [1984] AC 398 at 415 per Lord Keith. Back to Text.

14 See Auburn, `Usenet News and the law' [1995] 1 Web JCLI. Back to Text.

15 The Scotsman, 18 July 1997, cited in Edwards, n.2 supra. Back to Text.

16 See Butler, `Regulation of Content of On-Line Information Services- Can Technology Itself Solve the Problem it has Created?', (1996) 19(2) UNSW Law Jnl 193 at 201. Back to Text.

17 Senate select committee on Community Standards Relevant to the Supply of Services Utilising Electronic Technologies, Report on Video and Computer Games and Classification Issues, Oct, 1993 at 13. Back to Text.

18 ABA, Investigation into the Content of On-Line Services (December 1995). Back to Text.

19 See Koomen, `Illegal and Harmful Content on the Internet: Some Issues and Options', (April, 1998) No.35, 1. Back to Text.

20 See Flint, `Regulation of content on the Internet-an Australian Perspective', (1998) 1(9) TeleMedia 158 at 160. Back to Text.

21 See Graham, `Will PICS Torch Free speech on the Internet?', (1998) 17(1) Communications Law Bulletin 11. Back to Text.

22 Ibid, 6. Back to Text.

23 See Lambert, `Self-regulation v. Censorship- ISPs and Internet Content Legislation in Australia', (1997) 16(2) Communications Law Bulletin 3 at 4-5. Back to Text.

24 Eg see Argy, `Internet Censorship Proposals' [Internet] URL: http://www.msj.com.au/argy/intreg1.html. Back to Text.

25 Eg s.99 of the WA Censorship Act 1996 includes as `objectionable material' child pornography, articles promoting violence and involving bestiality, necrophilia, and torture. Back to Text.

26 Ibid. Back to Text.

27 Western Australia, 34th Parl., 4th Sess., Parliamentary Debates, Warnock, Back to Text.

28 eg see Francis, `Victorian Internet Censorship legislation- Is it Constitutionally Valid?', (1997) 16(2) Communications Law Bulletin 7. Back to Text.

29 [1990] HCA 24; (1991) 171 CLR 1. Back to Text.

30 [1997] HCA 25; (1997) 145 ALR 96. Back to Text.

31 See Richardson, `Freedom of Political Discussion and Intellectual Property Law in Australia' [1997] 11 EIPR 631. Back to Text.

32 [1988] HCA 63; (1988) 166 CLR 79. Back to Text.

33 Ibid., at 104 per Brennan J. Back to Text.

34 Cited n.1, supra. Back to Text.

35 Quoted in Corker and Sala, `The Supreme Court's Ruling on the Communications Decency Act: A Victory for Free Speech', (1997) 16(2) Communications Law Bulletin 1 at 3. Back to Text.

36 N.1, supra at 15. Back to Text.

37 Eg see Mcgee, `The Future of Free speech on the Internet', Computer Law Seminar, December 2, 1997, [Internet] URL: http://www.law.stetson.edu/courses/computerlaw/papers/jmcgeef97.htm. Back to Text.

38 See [Internet] URL:http://cdt.org/speech/empower.html. Back to Text.

39 Eg see Internet Family Empowerment White Paper, ibid. Back to Text.

40 See ACLU, `Fahrenheit 451.2: Is Cyberspace Burning- How Rating Proposals May Torch Free Speech on the Internet', [Internet] URL: http://www.aclu.org/issues/cyber/burning.html. Back to Text.

41 Ibid. Back to Text.

42 See eg the authorities cited in Weinrib, `Information as Property' (1988) U. Tor.L.J.117 at 120-123, but contrast Vaver, `Intellectual Property Today: Of Myths and Paradoxes' (1990) 69 Can. Bar Rev. 98; P. Drahos, A Philosophy of Intellectual Property, London: Dartmouth, 1996. Back to Text.

43 Adobe Systems Inc v Southern Software Inc (ND Cal 1998) noted in [1998] CTLR N-70. Back to Text.

44 Eg see Maltz, `Copyright, Hyperlinks & Networked Information Resources- Changing the Contours of Intellectual Property Law' (19970 16(1) Communications Law Bull. 1. Back to Text.

45 [1997] FSR 604. Back to Text.

46 Eg O'Mahony, `Web Issues', [Internet] URL: http://www.benedict.com/webiss.html. Back to Text.

47 Eg see Nelson, `Transcopyright: Pre-permission for Virtual Republishing' [Internet] URL: http://fizzgig.glasswings.com.au/xanadu/transcopy.html. Back to Text.

48 No 96C2966 ND Ill. Nov.13, 1997, cited in Ustaran, `Web Sites and Copyright Infringement' (June/July 1998), 32 at 33. Back to Text.

49 See (1997) 11 World intellectual Property Report 55. Back to Text.

50 839 F. Supp. 1552 (MD Fla, 1993). Back to Text.

51 See Loundy, `E-Law 2.0 Computer Information Systems Law and System Operator Liability Revisited' at n.373, [Internet] URL: http://www.austlii.edu.au/au/other/elaw/v1no3/loundy1.html. Back to Text.

52 [Internet] URL: http://www.wipo.org/eng/diplconf/distrib/94dc.htm. Back to Text.

53 Which, it seems, indicate an intention on the part of the Australian Government to accede to the Treaty, notwithstanding the fact that, at the time of writing it had not yet done so; nor had it acceded to the WIPO Performances and Phonograms Treaty: n 26 supra. Back to Text.

54 Commonwealth of Australia, Copyright Reform and the Digital Agenda: Proposed Transmission Right, right of Making Available and Enforcement Measures (July 1997) [Internet] URL: http://law.gov.au/publications/digital.htm. Back to Text.

55 Ibid., para 4.9. Back to Text.

56 Ibid., para 4.11. Back to Text.

57 Ibid., para 4.14. Back to Text.

58 High Court of Australia, unreported, 14 August 1997; [Internet] URL: http://www.austlii.edu.au/au/cases/cth/high_ct/unrep338.html. Back to Text.

59 Although Gummow J at first instance in the Federal Court of Australia sustained neither of APRA's arguments: see APRA v Telstra Corporation Ltd [1993] FCA 542; (1993) 118 ALR 684. Back to Text.

60 Note 4 supra, para 4.72. Back to Text.

61 Note 4 supra, para 4.71. Back to Text.

62 [1975] HCA 26; (1975) 133 CLR 1. Back to Text.

63 See also Macmillan and Blakeney, `The Internet and Communication Carrier's Copyright Liability', [1998] 20 EIPR 52. Back to Text.

64 Ibid., 20. Back to Text.

65 Cook, `Why Internet Service Providers Should be Copyright Guardians' (1996) 60 Copyright World 18. Back to Text.

66 Ibid., 19 Back to Text.

67 See, eg, Cook, n.61 supra; Loughnan, n 35 supra; Tickle, `The Vicarious Liability of Electronic Bulletin Board Operators for the Copyright Infringement Occurring on their Bulletin Boards' (1995) 80 Iowa Law Review 391; Haftke, `Net Liability: Is an Exemption from Liability for On-Line Service Providers Required?' [1996] 2 Entertainment Law Review 274; Hagen, `On-Line Service Provider Liability: The Latest US Copyright Conundrum' [1996] 7 Entertainment Law Review 274; Melone, `Contributory Liability for Access Providers: Solving the Conundrum Digitalization Has Placed on Copyright Laws' (1997) 49 Federal Communications Law Journal 491 [Internet] URL: http://www.law.indiana.edu/melone.html. Back to Text.

68 See Hails, `Liability of On-Line Service Providers Resulting from Copyright Infringement Performed by their Subscribers' [1996] 5 EIPR 304, citing Religious Technology Center v Netcom On-Line Communication Services Inc 907 F. Supp. 1361 (ND Cal., 1995). Back to Text.

B 69 ainbridge, `Internet Service Providers, Web Site Operators and Copyright Infringement' (1998) 14 Computer Law and Security Report 107 at 110. Back to Text.

70 See Blakeney, `Well Known Marks', [1994] EIPR 481. Back to Text.

71 Eg Roadrunner Computer Systems, Inc v Network Solutions Inc Back to Text.

72 [1998] FSR 16. Back to Text.

73 See Hurdle, `Domain Names - The Scope of a Trade Mark Proprietor's Monopoly Avnet Inc v Isoact Ltd', [1998] 2 EIPR 74. Back to Text.

74 Ibid. Back to Text.

75 See Meyer, `Intellectual Property Rights on the Internet', (1998) 14 Computer Law & Security Report 14 at 17. Back to Text.

76 Unreported, discussed in Osborne, `Domain Names Registration & Recent Resolution and Recent UK cases', [1997] 11 EIPR 644 at 646. Back to Text.

77 See Stoodley, n.4 supra, at 511. Back to Text.

78 See WIPO, Resolution of Intellectual property disputes within the context of the Memorandum of Understanding on the Generic Top-level Domain Name Space of the Internet Domain Name System (gTLD-MoU) TDN/CM/1/3, 30 May 1997. Back to Text.

79 [Internet] URL: http://www.law.uoknor.edu/hist/state97html cited in McGee, n.37 supra . Back to Text.

80 Cited n.1 supra. Back to Text.

81 Ibid., 10. Back to Text.

82 Eg see Urbancic v Shortland County Council. Back to Text.