Skip to main content Skip to navigation

JILT 1998 (2) - P. Eve Athanasekou

Internet and Copyright:

An Introduction to Caching, Linking and Framing

P. Eve Athanasekou
The School of Law
University of Glasgow

1. Introduction
2. Caching
3. Linking
4. Framing
5. Recommendation

Word icon and download article in .doc format Download


The interaction of copyright and the Internet has resulted in several legal quandaries concerning some of its technical characteristics. The following article is an introduction to the basic copyright issues in relation to the technical characteristics of caching, linking and framing.

Keywords: Internet, World Wide Web, copyright, caching, linking, framing.

This is a Work in Progress article published on 30 June 1998.

Citation: Athanasekou P E, 'Internet and Copyright: An Introduction to Caching, Linking and Framing', Work in Progress, 1998 (2) The Journal of Information, Law and Technology (JILT). <>. New citation as at 1/1/04: <>.

1. Introduction

One of the most significant questions surrounding the emerging Internet technologies is how copyright can be applied in light of the ease with which vast amounts of data can be stored, copied, manipulated and transmitted.[1] The resulting debate has revealed that there is a large number of jurists and cyberspace theoreticians who seriously contest [2] the suitability of the existing regime to regulate material on the Internet, and call for its abolition or its revision.

The problems stemming from the interaction of the Internet technologies and copyright law span several different categories (legal, sociological, technological). Below, I will attempt to give a brief introduction to the legal situations created by the technological characteristics of the Internet. These characteristics create new types of legal quandaries that the pre-existing law could not have predicted and, subsequently, cannot regulate effectively. The three most important ones are:

  • Caching
  • Linking and
  • Framing.

2. Caching

'Caching' is a generic term, which refers to the process of making an extra copy of a file or set of files for more convenient retrieval. On the Internet, caching occurs both on the user's computer (transient RAM copies) and at server level ('proxy caching').[3] The creation of transient copies on the RAM memory of a computer has been dealt with in the context of software copyright law as an infringement, but since this is the way that computers work, it can be covered quite easily by the concept of 'fear dealing'.

Proxy caching, on the other hand, requires a more sophisticated legal analysis, because it involves file duplication on a much larger scale. 'Caches', the servers that hold the copies of Internet files, have an almost unlimited storing capacity, in order to facilitate easy access to users by minimising access time to files. Without caching, the already heavily taxed Internet infrastructure would be unable to cope with user demands, and would finally collapse.

Despite, however, its significance for the continuous use of the Internet, caching has a profound impact on copyrighted material. Not only it facilitates the copying of entire web sites, but it also affects revenue for the site owners by delivering out-dated copies, disguising the true number of hits and visits, and reducing the effectiveness of advertising. A fair dealing defence is, therefore, not as straightforward as in the case of RAM transient copies, depending not only on the functionality of 'proxy caching', but also on the type of the cached web sites (commercial, non-commercial). Nevertheless, ubiquitous practice is about to create a de facto situation, since there is a tendency for the creation of bigger and bigger caches for faster navigation.

3. Linking

The most popular of Internet tools, the World Wide Web, operates in a non-linear fashion, which is achieved by means of cross-referenced links to associated items of information. This associational browsing mode, known as 'linking' (or 'hyperlinking'), is a unique characteristic of the World Wide Web, and it is considered the main reason for its popularity, since it facilitates effective navigation without requiring extensive skill or knowledge from the user.[4]

Links are references to Internet addresses (or Uniform Resource Locators - URLs). They are easy to spot on any webpage, since they usually appear as line(s) of blue-coloured (often underlined) text, which, when selected by a user, enable transition from one document to another (on the same or a different server) or connect different parts of the same document. As mentioned above, links are references to Internet addresses, i.e. statements of locations without original content, which fall under the category of facts. As such, they cannot be copyrightable according to the 1988 Copyright, Designs and Patents Act (CDPA) or any other copyright legislation, for that matter.

The past couple of years have witnessed claims that linking infringes copyright and that it should not be permitted. The problem stems from a misunderstanding of what a link does. Given the ease, with which material is retrieved through links, a lot of people confuse them with actual content. Using a link does not cause any copy of a web page to be created on the user's computer or the web author's server. Furthermore, it does not stake ownership claims on the linked site by the person making the link. Linking is more or less similar to placing references to other people's work. In fact, it has been argued that by placing their materials on the Internet (without stating expressly their objection to external links), copyright owners grant 'implied licences' to other users to link to them.[5] As the founder of the World Wide Web, Tim Berners-Lee[6] points out, there is no reason to have to ask for permission[7] to link to other documents. Not only it comports well with common practice, but also with the rationale behind the creation of the World Wide Web as a repository of knowledge. Banning the use of links could have profound impact on access to the Internet, since locating information would be extremely difficult.

In late 1996, links became for the first time the cause of controversy with regard to copyright in a dispute originating from the Shetland Islands in Scotland. The Shetland Times Ltd v. Dr Jonathan Wills and ZetNews Ltd[8] has received widespread[9] comment in the UK and international press. Both parties offered Internet-based news services. In October 1996, "Shetland News" reproduced verbatim a number of headlines from the online edition of the Shetland Times as hypertext links to the corresponding news articles.[10] This resulted to users bypassing the front page of the Shetland Times' web site (which contain all the advertisements) and reading only the linked texts. When the case came before the Court of Session (Edinburgh), the pursuers argued that both their web site and that of the defenders were cable programmes (within the meaning of s7 the CDPA 1988), and that the defenders infringed copyright under s20 of the Act (infringement by broadcasting or inclusion in a cable programme service). Granting an interim interdict (interlocutory injunction), the judge ordered that until the final judgment, all links to the "Shetland Times" from the defenders' web site should be removed. There is nothing in the interdict suggesting that copyright can be employed to prevent the use of hypertext links from one site to another, and the legal question was not whether links infringe copyright, but whether headlines are literary works and deserve copyright protection. Many commentators,[11] however, have taken the view that this preliminary decision represents an attack on the freedom to link associated with the World Wide Web, basing their comments mostly on the two parties' statements. The Shetland News case settled[12] out of court in November 1997, so there is no final judgment that can be used as a legal precedent for future linking cases.

4. Framing

'Framing' (or 'frame-linking') is another common practice on the World Wide Web, although not as ubiquitous as linking. The FRAME tag, which was conceived in early 1996 by Netscape Communications Corp. as a proprietary feature of their web browser, allows the viewer screen to split into multiple independently scrollable sections ('frames'), which may contain text, images, or other frames.

In their most common form, frames are used for the creation of static windows presenting title graphics, menus or tables of content. Typically, a small frame to the left is used as a menu, and a large one to the right contains the body of text. The materials presented within the frames usually originate from the same web site and content provider. When in this form, there is no legal controversy, since frames are no more than a tool for presenting information in a neater, more structured way. There are however, web sites that use frames to present information from some other web site surrounded by their own logos, or advertisements. This framing practice has led to disputes,[13] because the content providers, whose content appeared framed, considered that an infringement of their copyrighted material.

Despite the fact that, at first, the assumption of copyright infringement may seem justified, there are several issues that contest its accuracy. It is true that many users could be misled to believe that all material on screen originates from the same source, especially since the URL address displayed remains that of the framing site. However, no copying of files takes place on the intermediary web site and the material presented within the frame is still accessed through a link, which is no more than a hypertext reference. But, whereas in the case of simple links, we can argue that there is an implied licence for linking, that does not seem as easy to do in the case of frames, due to this confusion created by the URL display.

Unfortunately, and despite numerous lawsuits involving frames,[14] there are no fixed rules. All cases so far have ended in out-of-court settlements, whose terms have been dictated by the balance of powers between the parties rather than facts, or technology. A most characteristic example is that of the first framing case, Washington Post Co. v. Total News.[15] The defendant has a web site offering access to several news reporting sites, such as Washington Post, Dow Jones, Reuters, Time Mirror, etc. This is not the first site of the genre;[16] it differs, though, from the rest because it uses frames. The plaintiffs alleged, among other things, copyright and trademark infringement, trademark dilution, misappropriation, etc. The case ended with a settlement, according to which TotalNews undertook not to display the plaintiffs' stories within frames and was granted a 'linking licence,' which could be revoked anytime. The outcome of the case did not come as a surprise to anyone; TotalNews is a 5-person company that, even if they wanted to continue the dispute, they could not have afforded it. Nevertheless, other news providers, such as National Public Radio and MSNBC, whose sites have been also framed by TotalNews, have agreed to allow it to continue displaying their stories within frames, since its frame-links from brought them more visitors. TotalNews continues to use frames for all web sites in its 'personal edition,'[17] which allows the users to customise their preferences.

5. Recommendation

The above issues are merely examples of the problems, which can arise from new technologies. Most relevant disputes turn out to be semantic, a fact that shows how inappropriate the current interpretation of copyright law can be. There are technical means that can prevent web sites from being linked to with 'deep links' or 'framed,' and they are already being used by some site owners. There are several tools[18] for that, some as simple as a short line of code[19] that can be inserted to all web documents and force the browser to display the web site in full screen, or only its front page.

A proposal from a well-known Internet designer[20] summarises the problem in a very clear way; he says:

The potential negative effects of a bad precedent arising from a ruling by a judge ignorant of how the World Wide Web works are profound ... [I]nformation technology changes so fast [that]... the world would be better off, if there was a general rule requiring complainants in areas of fast-moving technology and murky law to exhaust all reasonable technological remedies before resorting to the legal system.


Athanasekou P.E 'Copyright in Cyberspace', 13th Annual BILETA Conference, March 1998 (Soon to be available from the BILETA site. <>

Berners-Lee T (1997) Links and Law: Myths <>

Bolin B and Tysver D (1998) Linking and Liability <>

Cyberlaw Institute (CLI) (1995) Copyright Law on the Internet: The Special Problem, of Caching and Copyright Protection <>

Finkelstein S (1997) 'The TotalNEWS Lawsuit', ILPN 4/97 <>.

Jun M (1997) 'Been 'Framed'?: Impostors Beware!', N.Y.L.J. 20/6/97 <>.

Kuester J and Nieves P (1998) 'Hyperlinks, Frames and Meta-Tags: An Intellectual Property Analysis', 38 IDEA 243 <>.

Lloyd I (1997) Information Technology Law (London: Butterworths)

Luria, M (1997) 'Controlling Web Advertising: Spamming, Linking, Framing, and Privacy', The Computer Lawyer, Vol. 14 No. 11 <>.

MacQueen H (1997) 'Copyright and the Internet' in Edwards L and Waelde C (eds) Law and the Internet - Regulating Cyberspace (Oxford: Hart Publishing)

Merrill C and Burger R (1996) Keeping the Chain Unbroken <>.

Oppendalh & Larson (1996) Web Law FAQ <>.

Oppenheim C (1996) The Internet Copyright Case and its implications for users of the WWW <>.

Poler, E (1997) 'Frames and License Agreements', ILPN 20/10/97 <>.

Post D (1995) 'New Wine, Old Bottles: The Case of the Evanescent Copy', American Lawyer <>.

Raysman R and Brown P (1997) 'Danger Liaisons: The Legal Risks of Linking Web Sites', N.Y.L.J. 8/5/97 <>.

Sanger L (1996) Caching on the Internet <>.

Shetland Times Ltd v. Jonathan Wills and Another [1997] SLT 669 <>.

Shetland Times Ltd v. Jonathan Wills and Another. Settlement <>.

Simpson M (1997) 'Hyperlink at Your Own Risk', N.Y.L.J. 24/6/97 <>.

Washington Post Co. v. Total News, 97 Cv. 1990 (PKL) (S.D.N.Y.) [The lawsuit is at <>].


[1] David Post, 'New Wine, Old Bottles: The Case of the Evanescent Copy', American Lawyer (May 1995) <>

[2] P. Eve Athanasekou, 'Copyright in Cyberspace', 1998 BILETA Proceedings, pp. 2-3 (Soon to be available from the BILETA site. <>

[3] See Lisa Sanger, Caching on the Internet (1996) <> for analysis of the distinction between 'client' and 'proxy caching'.

[4] Cf. para. 41 from Microsoft's Answer to Ticketmaster's Complaint in Ticketmaster Corp. v. Microsoft Corp. No. 97-3055 DDP (C.D. Cal. Apr. 12, 1997) <>, where it is stated that linking 'is an intended consequence of and fundamental to the nature of the World Wide Web.'

[5] Charles Oppenheim, The Internet Copyright Case and its implications for users of the WWW (Dec. 1996) <>; Ellen Poler, 'Frames and License Agreements', ILPN 20 Oct. 1997 <>.

[6] Tim Berners-Lee, Links and Law: Myths (April 1997) <>.

[7] Out of courtesy, one may ask permission to link, but this is not a Netiquette rule, i.e. part of the conduct code for Internet users. However, if you have any doubts about a particular link, then asking for permission seems a very reasonable thing to do. Furthermore, it may result to a reciprocal link from that page back to yours.

[8] Shetland Times Ltd v. Jonathan Wills and Another. Before Lord Hamilton [Judgment 24 Oct. 1996], 1997 SLT 669. For extensive comments, see Ian Lloyd, Information Technology Law (2nd ed. - 1997), pp. 359-364.

[9] Relevant articles appeared in the Times, the Irish Times, the New York Times, the Sonntagzeitung, the Christian Monitor, the New Scientist, etc.

[10] This type of links is called 'deep links', in order to be distinguished from links to the front page of the web site ('home page'), which are called 'surface links.'

[11] E.g. Bird Semple, Shetland Times -v- Shetland Times <>.

[12] The terms of the settlement are at <>, as well as at <>.

[13] Richard Raysman and Peter Brown, 'Danger Liaisons: The Legal Risks of Linking Web Sites', The New York Law Journal (8 April 1997) <>.

[14] Washington Post Co. v. Total News, Los Angeles Times v. Lycos, Los Angeles Times v. Free Republic, v. Austro.NET, etc.

[15] Washington Post Co. v. Total News, 97 Cv. 1990 (PKL) (S.D.N.Y.): The lawsuit is at <>. See also Meeka Jun, 'Been 'Framed'?: Impostors Beware!', The New York Law Journal (20 June 1997) <>.

[16] See for example CNET's NEWS.COM <>.

[17] Martha Stone, 'TotalNEWS takes it personal', ZDNet 10/06/1997 <> and 'Big Six: We' ve been framed, again!', ZDNet 11/06/1997 <>.

[18] E.g. Kristi Coale, 'Intellicast Smartens Up to Banner Bypass', Wired News 28 March 1997 <>. See also Jeffrey R. Kuester and Peter A. Nieves, 'Hyperlinks, Frames and Meta-Tags: An Intellectual Property Analysis', 38 IDEA 243ff. (1998) <> under V. Recommendations.

[19] The HTML Tag BODY can be 'enriched' to prevent framing. For more details, see Chuck Musciano, 'The best way to FRAME-proof your pages', Netscape World May 1997 <>. See also Web Design Group (WDG), Frames FAQ: Avoiding getting 'framed' <> for a JavaScript solution to the same problem.

[20] Seth Finkelstein, 'The TotalNEWS Lawsuit', ILPN April 1997 <>. Cf. 'Web Link Lawsuits Raise Serious Questions - Comments of the Electronic Frontier Foundation on Web Content Linkage Lawsuits', EFFector Vol. 10 No. 4 (17 March 1997) <>.

JILT logo and link to JILT homepage