article


Linking online content - enforcement of rights and hitting the right target


19 February 2010


Estimates vary, but it is commonly accepted that there are over 150 million websites on the web (and reportedly over 25 billion web pages). Without the ability to navigate logically and coherently between those websites and pages, the effective management of online content would have become impossible, and the web would have become uncontrollable, a long time ago.

Since the web is driven by hypertext mark-up language (html), the natural way to navigate around websites is through the use of hypertext links. Such links have become the day to day marketing, social networking, research and commercial tools of the online trade - the online signpost.

However, as soon as content providers direct their readers to third party websites the intellectual property rights of competing parties and obligations flowing from those rights come into play. Does a content provider automatically grant a non-exclusive licence to the world at large to link his content to other sites, simply because he has put his content on the web? If not, how can he regulate who accesses and redirects his content?

In recent times the debate has circled in particular around the liability of internet service providers and most recently search engines (which are, after all the custodians of a plethora of hyperlinks). Of those 150 million websites, it’s estimated that Google alone hosts around 11 million. To what extent should those entities be responsible for the links they provide, and the content of the third party websites to which they point?

What’s in a link?

A hyperlink is nothing more than an electronic address or Uniform Resource Locator (URL) which points to a location on the web. It can be a link to a page within the same site as where the link appears or to a different site. The link can be displayed in any number of ways, for example as the URL of the location it is pointing to or as a key word or image associated with that location.

Outlinking

The most often used link is the outlink. Put crudely, it works by storing the electronic address of the site to which it points. When activated by the user clicking on it, it sends the user to the destination page either within the same window which the user has opened or by opening the destination page in a new window.

The outlink comprises either a simple link (or surface link) which points to the homepage of a website, or a deep link, which points to a page (other than the homepage) within a new website. The user can determine that he has been redirected to a new page because the URL will have changed, but problems arise in particular where deep-linking is used. For example, has the user missed some important information on the homepage (some advertising, guidance or website terms and conditions)?

Inlinking and Framing

By contrast an “inlink” pulls content into the host site from a different page. This means that the user’s view of the current page is not interrupted, but rather additional data from a different page supplements that view. The user may be unaware that the material he is now viewing is from a third party website. The likelihood that the user will make certain assumptions in relation to the content is thereby increased. For example, what is the origin of the products he is now viewing?

Is linking an infringing act?

Copyright and database right issues

Firstly, it is unlikely that copyright subsists in a URL itself. Following Ticketmaster v Tickets.com in the District Court of Central California (6 March 2003) the position under US law is that a URL does not have the requisite quality of originality to attract copyright. It is difficult to see how the position would be different under English law. So providing a link which comprises the URL of a third party website is very unlikely to infringe copyright. Of course, if the link is a graphic or image which points to a third party page, there is a risk that reproduction of that graphic or image is an infringement of copyright, and so care must be taken when considering how to represent the link.

Providing a simple link from one site to another was until a few years ago considered to be unlikely to infringe copyright of that destination site. After all, creating a link does not require an act of copying of the destination site onto the server of the link provider. It is a signpost, or a reference, not a provision of content. The position of the end user is different of course, because he will be making a copy of the content of the destination site on his server.

However, with the implementation of the EU Copyright Directive (2001/29/EC) in the UK, section 20(2)(b) of the Copyright Designs and Patents Act 1988 (“CPDA”) was introduced. This provides an exclusive right of an owner of a copyright work to “communicate the work to the public” by electronic means “in such a way that members of the public may access it from a place and a time individually chosen by them”.

This means that a third party requires the authorisation of the copyright owner to communicate the work to the public. Without such authorisation, that third party infringes copyright. Arguably, this authorisation is given by the owner when he puts the content onto his own website (particularly if no restrictions as to reproduction accompany that content). After all, in a large number of cases the owner of a website will welcome more traffic on his site and so actively encourages others’ communication of the work to the public (think viral marketing).

However, a content provider will want to avail himself of the right to withhold his authorisation in certain circumstances, particularly where he wants to preserve the exclusivity of his content and strictly control any associations with third party websites. This is particularly the case where the content provider wants to protect his content from being deep linked to a third party website. It is not suggested that a deep link is any more likely to infringe copyright than a simple link. Again in the Ticketmaster case, it was held that deep linking did not amount to copyright infringement, but the position remains untested in the English courts following the introduction of section 20(2)(b) CPDA. Therefore a content provider wishing to restrict circulation of his website by others would be well advised to include the appropriate restrictions in his website terms and conditions or perhaps consider password protection of the website.

We should not forget that the content owner may have a right of action against an end-user of the content (i.e. the reader of the linked website). But even though the end user will have made a copy of the material on his own server by accessing the website, he may have a defence of fair use, unless he uses that material for commercial purposes.

By contrast, the sui generis right relating to databases has been effectively used to challenge those who deploy deep links to others’ websites. Courts have held that the content of a website can amount to a database, being a “collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means” (Article 1(2) of the Database Directive (96/9/EC). Providing a deep link to that website may constitute an extraction or reutilisation of a substantial part (or repeated and systematic extraction or reutilisation of insubstantial parts) of that database. In the case of Apis-Hristovich v Lakorda AD 5 March 2009, it was considered by the ECJ that the reproduction of hyperlinks within the database could itself be evidence of infringement.

Trade marks and passing off

The representation of a hyperlink may itself comprise an image, brand or logo, which carries with it the risk of trade mark infringement and passing off. Moreover, a content provider whose website is linked to a third party website may have a cause of action in trade mark infringement and passing off particularly if the linking carries with it a risk of confusion as to origin. This is more likely to arise in the case of deep linking since the user will not necessarily know the origin of the page to which he has been sent via the link, and also in the case of framing where he views the new content within the same window as the original content.

Reputation management

Although the act of linking itself is unlikely to carry with it a risk of an act of defamation or malicious falsehood (unless the link comprises a defamatory statement), the person providing a link must keep a keen eye on the content of the destination site and particularly the page to which he is directing the reader. If that destination page contains defamatory material then the act of linking provides an additional access point to that material and may be considered to be a republication of that material.

The risk is particularly high where defamatory material which has been created in one jurisdiction is accessed in a second jurisdiction. The provider of the link may find himself on the receiving end of a defamation action either in a foreign jurisdiction or in his home courts in respect of a foreign publication. So in Gutnik v Dow Jones, M3/2002 (May 28, 2002), the claimant sued in Australia (where most of his trade took place) even though the defamatory statement was published on the web in the United States.

Who’s liable?

Liability of search engines

There are a number of reasons why search engines have been targeted as defendants in copyright infringement actions in recent times. Publications and archives which are traditionally printed are looking to realise the value of their online archives and have in some cases moved to start charging for the provision of that online content. That market is however perceived to be threatened by search engines providing links to such content without charge, despite generating substantial advertising revenue. As a result the battle has raged between the press and the search engines, as to whether search engines can be liable as a result of the provision of such links.

Equally, the downloading of illegal MP3 files is facilitated greatly by increased accessibility. So, in a case originally brought in China by The International Federation of the Phonographic Industry on behalf of major record labels, search engine Yahoo has been held liable for providing deep links to illegal MP3 files on third party websites.

In the midst of this debate in the UK, Conservative Peer Lord Lucas had proposed an amendment to the Digital Economy Bill. This amendment, if adopted would appear to settle the matter, at least as far as the UK is concerned. The specific clause proposed to be inserted at section 116G of the CPDA read:

“Every provider of a publicly accessible website shall be presumed to give a standing and non-exclusive license to providers of search engine services to make a copy of some or all of the content of that website, for the purposes only of providing said search engine services…

A provider of search engine services who acts in accordance with this section shall not be liable for any breach of copyright….”

The presumption would be rebuttable by explicit evidence (only in the form of statements in a machine readable file – for example a text file blocking search spiders) that a licence was not being provided.

However, the amendment was not moved by Lord Lucas in committee stage. He explained that he had tabled them as probing amendments in order to determine the Government’s position on the subject. The Government has indicated that it would want to consult before bringing in such changes.

Quite how the Government will next move on this issue remains to be seen. Certainly those currently lobbying within the music industry for protection against illegal downloads will not welcome any move towards granting search engines what is perceived to be immunity from copyright infringement. Such immunity would pervade a large number of sectors looking to hold search engines responsible for the sites to which they direct their customers (not least the press).

Liability of internet service providers

Godfrey v Demon Internet Ltd [1999] 4 All ER 342 saw the advent of liability of internet service providers (ISPs) for the content they host. However, the implementation of the E-Commerce Directive (2000/31) by the Electronic Commerce (E.C. Directive) Regulations 2002 affords statutory protection to internet service providers which host or cache content.

In the case of ISPs caching information (i.e. storing information to effect an efficient retrieval of it) the ISP must:

a) store the information solely for the purpose of making more efficient onward transmission of the information to other recipients of the service;

b) not modify the information;

c) comply with conditions on access to that information and any rules regarding the updating of that information;

d) not interfere with the lawful use of technology to obtain data on the use of that information; and

e) act expeditiously to remove or disable access to that information upon actual knowledge that the information at the initial source of the transmission has been removed/disabled. (Regulation 18)

Likewise, the Electronic Commerce (E.C. Directive) Regulations 2002 deals directly with the liability of an ISP which hosts information, particularly in relation to liability in respect of unlawful content. The ISP is protected from liability if:

a) he does not have actual knowledge of the unlawful content;

b) if put on notice of that fact, acts expeditiously to remove or disable access to that content; and

c) provided the recipient of the service was not acting under the authority or control of the ISP.

This does, in fact, leave substantial scope for bringing ISPs into actions where defamatory content is uncovered. For this reason, many ISPs have sophisticated take-down policies and procedures.

Conclusions

When a content provider makes his content accessible to the online world via the web, he wants it to be visible. The CDPA provides him with the exclusive right to communicate the content to the public, but the strength of that right is doubtful, and arguably does not reflect the real world. Links to his content are to be expected, and for the most part encouraged. If the link really does misrepresent his content in some way (by associating it with content he considers to be undesirable) the content provider is likely to be able to take action. Meanwhile those in the serious and lucrative business of providing links around the web will continue to look over their shoulders in the UK, at least until the Government reports back on the Digital Economy Bill.

This article was first published in Copyright World

save to PDF

return to press office
return to press office
click here to return to the press office
more