The Ghost of Copyright Past – Collisions in the Digital Paradigm V

It is not too often that I have a personal experience of a Collision in the Digital Paradigm where the law becomes head to head with technological reality, but the following post is inspired by a recent occurrence.

When we were in London in late 2014 we saw a number of shows. We were spoilt for choice but one we wanted to see – and had missed on earlier occasions – was “Jersey Boys”, the musical about the Four Seasons. The experience in London was a good one. We walked from where we were staying at the East India Club to the heart of Piccadilly, had a quick meal and then on to the show. It was great – well staged, well acted, very slick and professional as one would expect. And then there was the music. That was always great but there is something extra special about hearing Sixties music with the sound technology that is available today. Other experiences include the music of the Beatles, remixed and remastered for the Cirque du Soleil show “Love” and the music of Elvis again for Cirque du Soleil’s “Viva Elvis”.  Those opening chords of “Jailhouse Rock” never sounded better. Both shows, incidentally, were at Las Vegas. But back to “Jersey Boys”. I missed the movie was well, so I was delighted to see that it was available on Bluray and DVD.

Time for a rewind. I have rather eclectic tastes in movies and have taken advantage of the DVD revolution (which would never have happened had the movie companies not been forced into the new market model of VHS [and subsequently DVD) as a result of format shifting and the case of Sony Corp of America v Universal City Studios 464 US 417 (1984)) to build up a collection. I encountered early problems with region coding. Many of the movies I wanted were not available in our Region 4 part of the world. It was necessary to obtain a region free DVD player or a software workaround for the DVD player on my computer to view DVDs sourced from Region 1 or 2 outlets. I had some concerns. The state of the law under our s.226 of the Copyright Act 1994 seemed to be ambivalent about circumventing technological protection measures like region coding where there were no infringement implications and the copy of the DVD had been legitimately acquired. There was a frisson of concern when the decision in Sony Computer Entertainment v Ball [2004] EWHC 1738 (Ch) was released with the suggestion in that case that content owners could set their own terms and conditions of sale of their product in addition to the statutory scheme of copyright – essentially allowing for copyright by contract. But as long as there was no copying there was no question of infringement.

With the passage of time more and more region-free DVD players became available and it seemed that distributor based market segmentation and regionalised markets were on the way out. Then along came Bluray and an even more rigorous form of region coding was instituted. Bluray players were not region free. And this was the position for a while. It was about 2102 that I heard of a region free Bluray player that was available in New Zealand, but it was pretty expensive. On the other hand it had a whole lot a features that made it very attractive. In addition, there were models available for a fraction of the cost of the NZ product but these could not be shipped to addresses outside the US. Fortunately NZ Post with its wonderful Youshop service with the provision of a US mailing address came to the rescue so welcome region free Bluray and DVD player with all sorts of additional goodies like 3D. The Bluray market became open slather.

And what about those copyright concerns. As a result of the 2008 amendments to the Copyright Act the definition of a technological protection measure does not include a device that controls access to a work for non-infringing purposes (for example, it does not include a process, treatment, mechanism, device, or system to the extent that it controls geographic market segmentation by preventing the playback in New Zealand of a non-infringing copy of a work) – see section 226 Copyright Act 1994. So circumventing access control mechanisms with  geographic implications – like region coding – is not viewed as any form of prohibited or infringing conduct. Thus, to circumvent a Region 1 (or A on Bluray) legitimately acquired disc to allow it to be viewed had no unlawful elements to it.

So back to the “Jersey Boys”. Having missed the movie and having enjoyed the stage show I purchased a Bluray version of the movie from Amazon. I addition to enjoying the show I was interested to see how Clint Eastwood made the transition from stage to screen. He is a fine director with some great credits. And the Bluray arrived. And inside was a little leaflet allowing me to add the movie to a Digital HD collection with Ultraviolet. This service, associated with an online retailer like Flixter enables access to movies in the collection on a number of devices. An advantage is that if one has a device like Chromecast a movie can be streamed via a computer or laptop and viewed on a TV screen. Great service and Chromecast is a great technology.

There is the leaflet with the code so that I can add “Jersey Boys” to my collection. I go to the Flixter site and enter the code, only to receive the message that the code is not valid for my region or territory. Once again it seems that region coding and market segmentation has hit the streaming video market. I am aware that this is a characteristic  or on-line content. Hulu and Netflix are available only to certain defined and regionally segmented IP addresses but it seemed to me that a legitimately acquired Bluray – that incidentally has NO region coding that was associated with an Ultraviolet code SHOULD be available world wide. I checked the fine print. Although I had obtained the copy from Amazon, there was a bold notice that the DVD was not for sale outside the USA or Canada. However, there was nothing on the Ultraviolet slip inside the case that suggested that the code was only valid for use inside the US or Canada.

I have always had difficulty with the idea of market segmentation. I have never supported the concept that content purchased from a particular geographical market could only be consumed within that geographical area. After all, a paperback book purchased in the US does not become “unreadable” as the plane leaves LAX. Why should DVDs or Blurays or Digital content be any different. The Internet has made markets global. The vendor from a suburb in Auckland who does business via the Web has a worldwide market – not just a local one. Similarly with those who sell and distribute on-line content. But the problem is that those most resistant to change in the Digital Paradigm are those who can benefit the most from it.

Why do I say that? Remember the case to which I referred earlier – Sony v Universal City Studios. That was a case where Sony, the developer of the Betamax videorecorder, sought relief from the US Supreme Court against the movie studios who sought to have the videorecorder banned because it was used for copyright infringement. Not so, held the USSC. There were substantial non-infringing uses, among them the concept of time shifting where one might record a show for viewing at a later time without engaging in infringement. The movie companies retired in disorder and then realised that the technology that they had tried to shut down could be used as an alternative means of movie distribution. In addition it meant that the back catalogue of the studios could be redistributed. This business model would have been lost if the Betamax case had been resolved in their favour. So the content distrbutors were – and to a degree still are – dragged kicking and screaming into the new Paradigm. DVD technology then replaced the videocasette and Bluray and streaming content will replace the DVD. But geographic segmentation remains the same – the ghost of copyright past.

When DVDs and digital content were in their early days, technological protection measures to guard against the wholesale copying of content were proposed. As Charles Clark said “the answer to the machine is in the machine”. In the same way there are work arounds to geographical restrictions – not only for region coded DVDs but also for streaming content. The legitimacy of such work arounds depends very much upon local law and the terms and conditions applicable to the product purchased.

It is not necessary to purchase the DVD or Bluray product to obtain the Ultra codes to add downloadable content to one’s library. Ultra codes are available for sale that allow this to be done. Once again the purchase of the code means that there are no infringement implications. So how does one avoid the “market segmentation” implications. As I say this depends on local law, but assuming that local law does not prohibit such workarounds, the use of a browser based anonymisation solution is probably the easiest and most user friendly way. Anonymox  is available both for Firefox and Chrome. The Firefox plugin works more effectively than the Chrome one, not the least because it showed the IP number and geographical location in the browser bar. But I emphasise, such a solution depends upon local law. It is not an offence in New Zealand to circumvent a TPM and a TPM does not include a device in place for market segmentation where the product has been legitimately acquired (although there may be contract implications). It is an offence under US law to circumvent a TPM. So one must be careful to comply with the law.

It is unfortunate that content owners are slow to recognise the new opportunities provided by the Digital Paradigm. It is unfortunate that content has not, until recently, been made available world-wide. This belated recognition now means that popular TV shows in the US are available within 24 hours of broadcast in the US – shows like “The Walking Dead”, “Arrow” “The Blacklist” and “Game of Thrones” come to mind. And what drove this? Piracy and content sharing. The content owners were reactive rather than proactive. And one can only wonder how long it will be before content owners wake up to the global market and get rid of this ridiculous hangover of region coding and market segmentation. It may have been understandable before the Internet Revolution. It has little if any relevance now.

But perhaps content owners prefer to adopt the well-known quote from Jean-Baptist Alphonse Karr – “plus ca change, plus c’est la meme chose.”

Linking and the Law – Part 2 – A Diversion to TPMs

Linking and the Law

PART 2

7    The New Zealand position — Technological Protection Measures, Anti-circumvention and communication

7.1             Introduction

The Digital Millennium Copyright Act came into force in the US in October 1998 as a response to the 1996 WIPO Copyright Treaty. Updated copyright legislation, including provisions relating to anti-circumvention of copy protection were enacted in the UK in 1988[54] and in New Zealand in 1994.[55]  In this section I shall consider the provisions of the Copyright Act that deal with Technological Protection Measures or TPMs.[56] It will become clear as the discussion progresses that the New Zealand legislation as amended by the Copyright (New Technologies)Amendment Act 2008 addresses the issues that were raised in Reimerdes and Corley.

The discussion in this section is admittedly lengthy but necessary to understand the approach to TPMs and the way that the Legislature has attempted to address  the problem. It is helpful within the context of linking because the decision in Reimerdes and Carley centreed around providing access to TPMs. A consideration of the New Zealand position, especially following the 2008 amendments to the Copyright Act will show significant differences in approach to TPMs from that in the DMCA and which would mean that the approach to linking in Reimerdes and Corley need not necessarily be applicable in New Zealand.

7.2             The Former Section 226 of the Copyright Act 1994

The provisions of the former s 226 of the Copyright Act 1994 created a right in favour of a person issuing copies (effectively a publisher). That person has the same rights as a copyright owner and the remedies that are available are provided.

Subsection (2) of s 226 defined how a person “infringes” the new right. The elements of the prohibited activity were:

•   Making, selling, offering or exposing for sale or hire; or

•   Advertising for sale or hire;

•   Any device or means;

•   Specifically designed or adapted to circumvent the form of copy protection employed; or

•   Publishing information;

•   With the intention to enable or assist persons to circumvent that form of copy protection;

•   Knowing or having reason to believe that the devices, means, or information will be used to make infringing copies.

State of mind is significant. For the publishing of information there were two states of mind involved. First, there had to be an intention to enable or assist persons to circumvent copy protection — the specific intention. Secondly, it had to be proven that the publisher knew, or had reason to believe, that the information would be used to make infringing copies.

The prohibition on the distribution of circumvention devices involved proof of the same state of mind relating to the use of devices. There had to be knowledge (or reason to believe) that the device would be used to make infringing copies. In addition, the device had to be specifically designed or adapted to circumvent the copy protection employed. Knowledge would seem to follow from the specific design or adaptation of the device. One would hardly distribute a circumvention device specifically designed for that purpose if one did not know or have reason to believe that the device would be used for circumvention purposes.

As far as devices or means were concerned it appeared that the use of those words extended not only to hardware devices that prevented copyright infringement taking place but the software devices such as DECSS.[57] As far as devices that have substantial non-infringing uses but incidentally include a circumvention device the situation is a little more difficult. At present DVD and Blu-Ray players have a built device that decrypts the CSS copy protection system. Imagine a DVD player/recorder that could not only play back material, but could record from a DVD as well. The CSS decoding system would be present for legitimate and authorised playback provisions. Thus the machine would be specifically designed to circumvent copy protection. But such a use would be authorised. Then there is the recording use. For liability to follow there would have to be specific knowledge on the part of the distributor of such a device that it would be used to make infringing copies. The mere presence of a circumvention means or device is not enough. It must be accompanied by the requisite knowledge or reason to believe.

The provision of information about circumvention means or devices was limited by two state of mind requirements that, arguably, would mean that the publication of, for example, academic research regarding circumvention technologies would not be caught by the section if:

•    the intention to enable or assist circumvention were absent; and/or

•    there was an absence of knowledge or reason to believe that the publication would be used to make infringing copies.

Thus, when we consider the examples the scope of the former section was somewhat narrower than it first appeared.

7.3              The 2008 Amendment

The 2008 Amendment of s 226 and following amendments have made a number of changes. The first is that definitions have been provided. The second is that the essence of the former s 226 is retained in s 226A.

The focus of the new s 226 continues to be on the link between circumvention and copyright infringement and on the making, sale and hire of devices or information rather than on the act of actual circumvention. Actual circumvention is not prohibited, but any unauthorised use of the material that is facilitated by circumvention continues to be an infringement of copyright.

The new amendments recognise that consumers should be able to make use of materials under the permitted acts, or view or execute a non-infringing copy of a work. This is consistent with New Zealand’s position on parallel importation of legitimate goods; for example, genuine DVDs from other jurisdictions. New provisions have also been introduced to enable the actual exercise of permitted acts where TPMs have been applied.

What the new TPM provisions do is two-fold — broadly they prohibit and criminalise.

There is a prohibition of commercial conduct that undermines the TPM by putting a circumvention device into circulation or providing a service including the publication of information which relates to overriding TPM protection. Contravention has civil consequences — specifically the issue of the work protected by a TPM is protected as if the conduct was an infringement of copyright. The second leg is to make the prohibited conduct a criminal offence.[58]

There is a knowledge element for both the prohibition and the offence — the knowledge of the use to which the circumvention device or the service or published information will, or is likely to, be put.

There are however some limits on the prohibition when circumvention device has a legitimate use.

7.3.1           The Definitions

There are three definitions which are applicable to ss 226A–226E. The first is a technological protection measure or TPM:[59]

TPM or technological protection measure

(a)    means any process, treatment, mechanism, device, or system that in the normal course of its operation prevents or inhibits the infringement of copyright in a TPM work; but

(b)   for the avoidance of doubt, does not include a process, treatment, mechanism, device, or system to the extent that, in the normal course of operation, it only controls any access to a work for non-infringing purposes (for example, it does not include a process, treatment, mechanism, device, or system to the extent that it controls geographic market segmentation by preventing the playback in New Zealand of a non-infringing copy of a work)

Significantly, the legislature differentiated between a TPM for the purposes of the prevention of infringement and one that relates to access to a work for non-infringing purposes. The example is given of the control of “geographic market segmentation”, which clearly relates to a region protection in games or DVDs. Thus, if a person legitimately acquired a DVD that was coded for region 1, the region coding device or process in the DVD player which would otherwise prevent the use of the DVD may be circumvented so that the non-infringing purpose of viewing the DVD could be carried out.

The second definition relates to a TPM circumvention device:[60]

TPM circumvention device means a device or means that—

(a)    is primarily designed, produced, or adapted for the purpose of enabling or facilitating the circumvention of a technological protection measure; and

(b)   has only limited commercially significant application except for its use in circumventing a technological protection measure

The primary purpose of the circumvention device must be to circumvent a TPM — taking into account that the TPM must prevent infringement rather than access for non-infringing purposes and as well as its primary design production or adaptation it must have limited commercially significant application other than for its use in circumventing a TPM.

Both paras (a) and (b) are conjunctive; it may well be that a TPM circumvention device may have other commercially significant applications or, as the Americans put it, substantial non-infringing uses.

The third definition relates to a TPM work which is defined as a copyright work that is protected by a TPM. A TPM work must be a copyright work but it may well be that this cannot prevent a entrepreneur locking up a public domain work with a TPM if there is some significance in the way in which the work has been typographically arranged.

7.3.2      The Operative Sections

Section 226A sets out the prohibited conduct in relation to a TPM, stating:

226A Prohibited conduct in relation to technological protection measure

(1) A person (A) must not make, import, sell, distribute, let for hire, offer or expose for sale or hire, or advertise for sale or hire, a TPM circumvention device that applies to a technological protection measure if A knows or has reason to believe that it will, or is likely to, be used to infringe copyright in a TPM work.

(2) A person (A) must not provide a service to another person (B) if—

(a)    A intends the service to enable or assist B to circumvent a technological protection measure; and

(b)   A knows or has reason to believe that the service will, or is likely to, be used to infringe copyright in a TPM work.

(3) A person (A) must not publish information enabling or assisting another person to circumvent a technological protection measure if A intends that the information will be used to infringe copyright in a TPM work.

Section 226A provides a useful example of modern statutory drafting techniques by clarifying the behaviours of the certain actors that the section addresses.

Section 226A(1) is identical in scope to the former s 226(1), with the exception that the definitions contained in the new s 226 impact upon the scope. Whereas the previous legislation referred to a form of copy protection, the definition of a TPM work, a TPM and a TPM circumvention device now govern.

Section 226A(2) relates to the publishing information limb of the former s 226, except that a new term (“service”) is used. This is undefined but clearly encompasses information.

Once again there are two limbs underlying the prohibition: the intention that the service enable or assists circumvention of TPM; and specific knowledge that the service will or is likely to be used to infringe copyright in a TPM work.

If the service is for the purposes of university research, it is difficult to imagine that B could be satisfied, thus the prohibitive conduct is not complete.

The use of the word “service” in s. 226A(2) is new. The earlier iteration used the words “device” or “means”. Service is a very wide concept and although s. 226A(3) refers to the publication of information to enable or assist another person to circumvent a TPM, service extends the scope of s. 226A(1) and in essence addresses any form of assistance enabling circumvention of a TPM accompanied by knowledge or reason to believe that the assistance or service will be used to infringe copyright. There seems to be little doubt that a “service” could concveivably encompass a computer program or code.

Section 226A(3) relates specifically to the publication of information, and although the behaviour could be encompassed by a service, the legislature saw fit to make publication of information about TPM circumvention a discreet behaviour.

There is only one knowledge element in s 226A(3), as opposed to the two in s 226A(2). That knowledge element is that person A must know that the provider of the information must intend that the information is to be used to infringe copyright in a TPM work. Thus s 226A prohibits:

•    the making or distribution of a TPM;

•    the provision of a service with the two limbs of intention to assist circumvention and knowledge that the service will be or likely to be used to circumvent; and

•    publication of information enabling circumvention if it is intended that information will be used to circumvent.

Unlike the original  s.226, which was restricted in the language to commercial activity (sells, lets for hire, offers or exposes for sale or hire or advertises for sale or hire) s. 226A prohibits not only the making, selling, letting for hire, offering or exposing for sake, but also prohibits importing or distributing a TPM circumvention device. These terms can encompass an individual who downloads a TPM circumvention device from an off-shore site. This was not thje case in the earlier legislation. Distibution is also prohibite3d. Thus if one makes a TPM circumvention device available for download from a website, and uses a link to facilitate delivery, such an action could fall within the ambit or “distribution”.

However, the provision of information has a commercial aspect to it, for the provision of such information must be “in the course of business” and is therefore of a narrower scope that had those words been omitted.

Section 226B sets out the rights that accrue to the issuer of a TPM work. These rights are what Kirby J referred to as para copyright in Stevens v Kabushiki Kaisha Sony Computer Entertainment.[61] Essentially, the issue of a TPM work has the same rights against a person who contravenes s 226A as the copyright owner has in respect of infringement. The provisions of the Copyright Act relating to delivery up in civil or criminal proceedings is available to the issuer of a TPM work as are certain presumptions that are contained in ss 126–129 of the Copyright Act. The provisions of s 134 relating to disposing of infringing copies or objects applies as well with the necessary modifications.

Absent from the 1994 version of s 226 was the offence of contravening s 226A. Section 226C creates that offence:

226C Offence of contravening section 226A

(1) A person (A) commits an offence who, in the course of business, makes, imports, sells, distributes, lets for hire, offers or exposes for sale or hire, or advertises for sale or hire, a TPM circumvention device that applies to a technological protection measure if A knows that it will, or is likely to, be used to infringe copyright in a TPM work.

(2) A person (A) commits an offence who, in the course of business, provides a service to another person (B) if—

(a)    A intends the service to enable or assist B to circumvent a technological protection measure; and

(b)   A knows that the service will, or is likely to, be used to infringe copyright in a TPM work.

(3) A person (A) commits an offence who, in the course of business, publishes information enabling or assisting another person to circumvent a technological protection measure if A intends that the information will be used to infringe copyright in a TPM work.

(4) A person who commits an offence under this section is liable on conviction on indictment to a fine not exceeding $150,000 or a term of imprisonment not exceeding 5 years or both.

The first important thing to note is that subs (4) requires the conviction to be on indictment, so the matter must be dealt with in the jury jurisdiction and cannot be dealt with summarily although the position may well be altered by the provisions of the Criminal Procedure Act 2011..

Section 226C mirrors the prohibitions in s 226A, but the critical matter for an offence is that there is a commercial element — “in the course of business”.

Similarly, the provision of the service in subs (2) of 226C must have a commercial element as must the publication of information in subs (3).

This then brings the criminalisation of para-copyright in line with the provisions of s 135 of the Copyright Act which relates to piracy or commercial infringement. Clearly, s 226D considers that the offence should relate to commercial activity involving TPMs. In this way the rather wider prohibitions contained in s. 226A  do not automatically lead to potential liability under s. 226C

Section 226D clarifies the position relating to the scope of the rights of the issuer of a TPM work. The operative part states:

226D When rights of issuer of TPM work do not apply

(1) The rights that the issuer of a TPM work has under section 226B do not prevent or restrict the exercise of a permitted act.

(2) The rights that the issuer of a TPM work has under section 226B do not prevent or restrict the making, importation, sale, or letting for hire of a TPM circumvention device to enable—

(a)    a qualified person to exercise a permitted act under Part 3 using a TPM circumvention device on behalf of the user of a TPM work; or

(b)   a person referred to in section 226E(3) to undertake encryption research.

(3) In this section and in section 226E, qualified person means—

(a)    the librarian of a prescribed library; or

(b)   the archivist of an archive; or

(c)    an educational establishment; or

(d)   any other person specified by the Governor-General by Order in Council on the recommendation of the Minister.

(4) A qualified person must not be supplied with a TPM circumvention device on behalf of a user unless the qualified person has first made a declaration to the supplier in the prescribed form.

The issuer of a TPM work cannot prevent or restrict the exercise of a permitted act. Nor can the prohibition prevent or restrict the making, importation, sale or letting for hire of a TPM circumvention device to enable encryption research under s 226E(3), or to enable a qualified person to exercise a permitted act using a TPM circumvention device.

The legislation goes on to define “qualified person”, who, in this case, is required to make a declaration relating to certain matters.

On their own the provisions of s 226D seem confusing, although the provisions of s 226 and following do not prohibit the act of circumvention. Subsection (1) of 226D makes it clear that circumvention may be permissible for the purposes of the exercise of a permitted act.

Section 226E takes the matter further.

226E User’s options if prevented from exercising permitted act by TPM

(1) Nothing in this Act prevents any person from using a TPM circumvention device to exercise a permitted act under Part 3.

(2) The user of a TPM work who wishes to exercise a permitted act under Part 3 but cannot practically do so because of a TPM may do either or both of the following:

(a)    apply to the copyright owner or the exclusive licensee for assistance enabling the user to exercise the permitted act:

(b)   engage a qualified person (see section 226D(3)) to exercise the permitted act on the user’s behalf using a TPM circumvention device, but only if the copyright owner or the exclusive licensee has refused the user’s request for assistance or has failed to respond to it within a reasonable time.

(3) Nothing in this Act prevents any person from using a TPM circumvention device to undertake encryption research if that person—

(a)    is either—

(i)     engaged in a course of study at an educational establishment in the field of encryption technology; or

(ii)    employed, trained, or experienced in the field of encryption technology; and

(b)   has either—

(i)     obtained permission from the copyright owner or exclusive licensee of the copyright to the use of a TPM circumvention device for the purpose of the research; or

(ii)    has taken, or will take, all reasonable steps to obtain that permission.

(4) A qualified person who exercises a permitted act on behalf of the user of a TPM work must not charge the user more than a sum consisting of the total of the cost of the provision of the service and a reasonable contribution to the qualified person’s general expenses.

Once again the section makes it clear that the act of circumvention to exercise a permitted act is not prohibited. Thus a person may use a circumvention device to copy a selection from a TPM work for the purposes of review, a comment or inclusion (with attribution) in an academic work.

Subsection (2) glosses over that, however. If the user of a TPM work wishes to exercise a permitted act, he or she may use a TPM circumvention device to do so, but the subsection includes the words “but cannot practically do so because of a TPM”. It is unclear what this means. If a person whose access to a work to carry out a permitted act is prevented by a TPM, does subs (2) automatically apply? Or, if a circumvention device is available, is the user able to use that circumvention device to exercise the permitted act? Does subs (2) relate to the situation where there is no circumvention device available? Subsection (2), in providing certain options for the person who is stymied by a TPM, challenges the market failure theory of fair use.

A person wishing to do one of the permitted acts may apply to the copyright owner or licensee for assistance. The alternative is to engage a qualified person (see s 226D(3)) to exercise a permitted act on the user’s behalf using a circumvention device. But that can only apply if the copyright owner exclusive licensee refuses the user’s request for assistance or fail to respond within a reasonable time.

A sensible interpretation of s 226E suggests that subs (2) must be followed if there is no readily available circumvention device enabling the user to exercise a permitted act.

It is also important to note that s 226E makes a specific exception for the use of circumvention devices to undertake encryption research in certain circumstances.

7.4              Comment

The new provisions of s 226 and following are indeed helpful. The incorporation of clear definitions, that make it clear that TPMs are for the purposes of prevention of infringement rather than access, are to be welcomed (although s 226E seems to introduce a somewhat unnecessary level of complexity).

Underlying the whole issue of para-copyright is the fact that, in reality, TPMs are a somewhat blunt instrument for the purposes of copyright protection, presenting an “all or nothing” level of protection. TPMs cannot discriminate between a permitted or prohibited use. They are international and are applied internationally, whereas copyright law is territorial. TPMs place the control in the hands of the copyright owner of a technological rather than a legal nature and, as already observed, provide a potential for market failure. Essentially, TPMs do not provide an absolute protection, rather they impose another layer of protection that sits on top of the balance of interests created by statute, and muddy the waters between what is and is not allowed. The various options relating to behaviour regarding TPMs contained in ss 226B, 226D and 226E suggest that certain behaviours may be permissible while others are not. Clearly, the legislature did not want to impose a prohibition on the act of circumvention, but the various alternatives given in ss 226D and 226E seem to suggest prohibition.

The legislation, while addressing the issue of circumvention of TPMs, and restricting prohibited conduct to the means by which copyright protection (rather than access prevention) may be circumvented therefore makes it clear that the provisions of means by which access controls may be circumvented is not within the scope of prohibited conduct. This means that one may provide services, information and programs that assist in circumventing access protections. In this way the legislation addresses its target – the copy right – rather than allowing the engraftment of another “para-copyright” – the “prevention of access” right. This is eminently justifiable. Region coding is a means by which copyright owners facilitate distribution of their products. The only issue is obne of market segmentation and a release strategy that copyright owners may have in place. There is no reason, in terms of copyright, why a person who legitimately acquires content in one geographical area should be prohibited from accessing it in another.

However, unlike the New Zealand legislation the DMCA prohibits thje circumvention of access control systems, despite there being no copyright implications and, to further complicate matters, criminalises such behaviour.  It should be a matter of concern that should international trade treaty negotiations result in the application of a DMCA style of anti-TPM circumvention regime, the results will be:

a) the imposition of a foreign marketing system that goes far beyond those chosen say for the release of non-digital product such as movies and CDs

b) the end of the parallel importing regime insofar as geographically segmented digital product is concerned

c) the criminalisation of behaviour that has nothing to do with copyright infringement and has no economic implications for content owners whatsoever.

Finally, it is still not clear whether licence terms or conditions of sale may override the way in which circumvention devices may be used in the limited situations provided in s 226. Unlike s 84, which statutorily negates such conditions, the matter is left open. The legislature has gone to considerable lengths to ensure the balance of interests that underlies copyright law is maintained. It seems unusual that those rights may be subverted by contractual arrangements.


[54]      The Copyright, Designs and Patents Act 1988.

[55]      The Copyright Act 1994 as amended by the Copyright (New Technologies) Amendment Act 2008

[56]      Section 226 – 226E

[57]      CSS is the DVD content scrambling system that prohibits the copying of the files on a DVD movie disk. DECSS is the system that circumvents the content scrambling system.

[58]      The offence of contravening s 226A is set out in s 226C.

[59]      See the new s 226.

[60]      See the new s 226.

[61]      Stevens v Kabushiki Kaisha Sony Computer Entertainment [2005] HCA 58, (2005) 224 CLR 193, (2005) 221 ALR 448.

Linking and the Law – Part 1

LINKING AND THE LAW

Hypertext Linking – Communication or Content Neutrality

Contents

PART 1.

1.            General Introduction

An Overview of Linking and Early Developments

2.            Introduction

3.            What is Linking

3.1          Linking

3.2          Deep Linking

3.3          Framing

3.4          In-lining

4.            Potential Copyright Issues Arising from Linking – the big picture

5.            The Cases

Shetland Times

The Washington Post Co & Ors v Total News Inc & Ors

Futuredontics Inc v Applied Anagramic Inc

Ticketmaster Corp v Microsoft

Ticketmaster Corp v Tickets.com

6.            Linking, Anti-circumvention and Copyright

6.1          Universal City Studios v Reimerdes and Corley

6.1.1      Background

6.2          The Categorisation of Linking Activity: The Decision in Universal City Studios v Reimerdes

Interpretation

The Real-World Analogy

Extending the Scope

Links – Kaplan J’s Approach

First Amendment, Jurisdictional and Technical Issues

1. Content-based or content neutral

2. Navigation on the Web

3. The Content Neutrality of Links?

4. Extraterritorial Effect of the Decision

5. Value Judgements and Technological Misunderstandings

6.3          Universal City Studios v Reimerdes and Corley – The Appeal

PART 2

7.            The New Zealand Position – Technological Protection Measures, Anti-circumvention and communication

7.1          Introduction

7.2          The Former Section 226 of the Copyright Act 1994

7.3          The 2008 Amendment

7.3.1      The Definitions

7.3.2      The Operative Sections

7.4          Comment

PART 3

8.            Linking and Publication – Ramifications for Defamation

8.1          Crookes v Newton

8.2          Concluding Thoughts on Crookes v Newton

9.            The European View –  Svensson v Retreiver Sverige AB

10.          Perfect 10 v Google – Moving Away from Reimerdes v Corley

11.          Linking – Issues Arising

12.          The Threat to the Web

13.          Conclusion

PART 1

1.               General Introduction

This is a lengthy discussion about hypertext linking and the law. It builds on earlier published pieces from the early 2000’s and a section that appeared in the 2nd edition of my text on internet law[1] but which is not available in the 3rd edition.

The revision and rewriting of the earlier material was prompted by some recent developments in the field including the decision of the Supreme Court of Canada in the case of Crookes v Newton [2] – a defamation case – and Perfect 10 v Google[3] – a copyright case, together with some recent European developments and the submission of the European Copyright Society in Svensson v Retreiver Sverige AB.[4] I delayed posting until the Court of Justice of the EU decision in Svensson was available. That decision has now been delivered and will be discussed in Part 3.

 I have divided the discussion into three parts which will be posted separately over a period of time.

Part 1 looks at the early cases on linking and in particular considers links to and the use of the material of others within the context of copyright. The discussion moves then to a consideration of technological protection measures (TPM) and anti-circumvention provisions provided bgy law – a concept described as “para-copyright” by Kirby J – and considers whether the provision of links to circumvention devices amounts to a breach of the Digital Millenium Copyright Act (US). The discussion focuses upon the case of Universal City Studios v Reimerdes and Corley[5] decided in 2000.

Part 2 considers the New Zealand approach to TPMs and whether the existing law would prevent linking to sites where TPMs are available and, if so, under what circumstances. The impact of the amendments to the Copyright Act 1994 incorporated by the Copyright (New Technologies) Amendment Act 2008 will be considered.  This is a new discussion even although the legislation has been in place for some time. There have been no cases on the New Zealand provisions.[6]

Part 3 is also new and considers the Canadian case of Crookes v Newton as well as approaches in various European jurisdictions together with the case of Perfect 10 v Google. I then conclude with some observations on some common themes that may be present about the state of the law regarding linking.

A Note on References

My penchant for references may be unusual in the context of a blog. Nevertheless I consider that to be useful and to provide evidential support for material that I post, references are as important here as they would be in an academic journal. I would hope that by providing references I may encourage other research on the topic or informed debate about the particular post.

This piece commenced life as a single article. It is too large for a single post. For that reason I have divided it into three parts although it is designed to be considered as a whole. The footnote references are sequential across the three parts, rather than starting afresh at the beginning of each part.

I have also departed from the strictures imposed by the New Zealand Law Style Guide.[7] Because this piece is spread across three parts, references to “above n. **” would be unhelpful and so I have repeated citations except when use of “ibid” is appropriate.

An Overview of Linking and Early Developments

2.               Introduction

In Universal City Studios v Reimerdes[8] the issues facing Judge Lewis Kaplan were whether or not the plaintiff was entitled to injunctive relief against the defendant to prevent:-

•    the defendant posting a copy of a DVD decryption program known as DeCSS on its website for distribution;

•    the defendant providing links from its website to other websites maintained by other persons, and which made DeCSS available.

The Judge held that the provision of links to other websites where the decryption program was available was the “functional equivalent” of making the program available itself.

The ramifications of Judge Kaplan’s decision go far beyond the realm of copyright and may impact upon the nature of the internet itself. This article examines the nature of linking and looks at some general principles of copyright and linking. Some of the cases involving linking and framing are discussed, and some of the more significant issues regarding linking are examined. The case of Universal City Studios v Reimerdes as decided at first instance by Judge Kaplan and the judgment of the Second Circuit Court of Appeal is considered. There is also consideration of whether or not the application of copyright principles to linking in fact presents us with a form of internet governance or regulation that could have the effect of suppressing the free exchange of information and content on the internet, and could possibly retard the development of the internet in the realm of ecommerce.

I also consider more recent cases about linking, including Perfect 10 v Google[9] which held that whilst links may have no implications for direct copyright infringement, the situation may be different for contributory infringement, as concept that has developed within United States copyright jurisprudence.

Although the copyright environment seems to dominate problems with linking the law of defamation also examines linking and the nature of publication in the new environment. I shall consider the Canadian case of Crookes v Newton[10] together with the New Zealand case of International Telephone Link Pty Ltd v IDG Communications Ltd.[11] It would be tempting to consider the recent decisions in Tamiz v Google,[12] A v Google,[13] Rana v Google[14] and Trkulja v Google[15] but these cases deal with the way in which search results are presented and whether a search engine becomes a publisher. They do not deal specifically with linking. But the line of cases may well be the subject of a separate article.

I shall conclude with an attempt to distil some common rules and themes from the cases, and consider whether the law is trying to apply old concepts that may no longer be valid in the new paradigm.

3                 What is linking?

3.1             Linking

Links or hypertext links are components of webpages. They allow one webpage, or a part of a website to connect to another webpage, either on the same or upon a different site. Clicking on relevant highlighted text or an image activates the link. Linking is seen as an essential navigation feature for moving around material on the world wide web.

3.2             Deep linking

A deep link is a hypertext link that takes a web user past a starting or top-level home page and directly to an internal page within a particular website. Deep linking can only be achieved if the internal document has its own Uniform Resource Locator field (URL).

3.3             Framing

Framing occurs when one website allows a user to view the contents of another website within a frame on the first website. In many cases the information is presented so that it appears to originate from the first website and to be a part of it. In fact this is not the case.

3.4             In-lining

In-lining is the process whereby graphic images originate from one site and are displayed on another. The html coding on the displaying site creates an unseen hypertext link to the site hosting the graphic image, thus creating the illusion that the image originates from the displaying site.

4.               Potential copyright issues arising from linking — the big picture

Some general issues arise from linking, especially in the area of copyright, that involve some fundamental aspects of computer use on the internet. Like it or not, when a website is accessed by a browser, the material from that website is stored in the RAM of the browser’s computer.

In addition, depending on the program that the browser uses (Internet Explorer, Chrome, Firefoxand so on), information regarding the site visited, the URL, certain images and the like may be retained by the program in the cache or the history sections (or both) of the program.

It is clear that those who place content upon the world wide web must be aware of this. Indeed, the reason why individuals and companies place material on the web is so that it may be accessed by others. Thus, there is an implied licence granted by the website owner to others so that they may view the material on their computers. That licence, it is suggested, is available by the placement of material on the web as such. However, if the browser utilises the material after viewing it, there may be an infringement issue. The real issue is how the browser deals with the material that has been accessed, to determine issues of copyright infringement.

It has been suggested that one can look at the circumstances surrounding the publication or distribution of the material, which would necessarily take into account the factors to which reference has already been made.[16] Campbell J took a more restrictive view of the matter in Intellectual Reserve Inc v Utah Lighthouse Ministry Inc where he said:

When a person browses a website, and by so doing displays the Handbook, a copy of the Handbook is made in the computer’s random access memory (RAM), to permit viewing of the material. And in making a copy, even a temporary one, the person who browsed infringes the copyright. See MAI Systems Corp v Peak Computer, Inc, 991 F 2d 511, 518 (9th Cir 1993) (holding that when material is transferred to a computer’s RAM, copying has occurred; in the absence of ownership of the copyright or express permission by licence, such an act constitutes copyright infringement); Marobie-Fl., Inc v National Ass’n of Fire Equip Distrib, 983 F Supp 1167, 1179 (N D Ill 1997) (noting that liability for copyright infringement is with the persons who cause the display or distribution of the infringing material onto their computer); see also Nimmer on Copyright § 8.08(A)(1) (stating that the infringing act of copying may occur from “loading the copyrighted material . . . into the computer’s random access memory (RAM)”).[17]

In a footnote the Judge observed that:

[T]he Copyright Act has provided a safeguard for innocent infringers. Where the infringer “was not aware and had no reason to believe that his or her acts constituted an infringement of copyright, the Court in its discretion may reduce the award of statutory damages” 17 U.S.C. § 504(c)(2).

However, it should be noted that this case dealt with a link that took users not to a page created by another web user but to a church handbook of instructions. Thus, what was downloaded was an electronic version of a book. Copyright law does not distinguish between a webpage and a book in electronic form that is available on the web. To copy either can constitute an infringement. However, it is suggested that Campbell J’s view cannot apply to all of the material on the web. Logically, the concept of an implied licence is applicable to a webpage placed on a server that may be accessed by any computer user.

The position regarding placing of a hypertext link on one page that takes a user to another page is less clear, and depends on the extent of the implied licence. The destination of the link has been the subject of litigation that is discussed in the next section. However, the Intellectual Reserve case has a certain relevance to this discussion. In that case the defendants had no objection to the injunction sought by the plaintiffs in so far as prevention of direct infringement was concerned. However, they resisted the injunction as to contributory infringement.[18] The defendants had placed a notice that the church handbook was online and gave three website addresses of sites that contained the material that the defendants were required to remove from their website.[19] The Court considered the nature of contributory infringement and found:

Liability for contributory infringement is imposed when “one who, with knowledge of the infringing activity, induces, causes or materially contributes to the infringing conduct of another.” Gershwin Publ’g Corp v Columbia Artists Mgt, Inc, 443 F 2d 1159, 1162 (2d Cir 1971). Thus, to prevail on its claim of contributory infringement, plaintiff must first be able to establish that the conduct defendants allegedly aided or encouraged could amount to infringement. See Subafilms, Ltd v MGM-Pathe Comms Co, 24 F 3d 1088, 1092 (9th Cir 1994). Defendants argue that they have not contributed to copyright infringement by those who posted the Handbook on websites nor by those who browsed the websites on their computers.

The Court in Intellectual Reserve found that those internet users who browsed the handbook on the three other sites infringed copyright on the basis observed above. The Court went on to find that the defendants induced, caused or materially contributed to the infringement in the following respects:

1.   After being ordered to remove the handbook from their website, the defendants posted on their website: “Church Handbook of Instructions is back online!” and listed the three website addresses.

2.   The defendants also posted emails suggesting that the lawsuit against them would be affected by people logging onto one of the websites and downloading the complete handbook.

3.   One of the emails posted by the defendants mentioned sending a copy of the copyrighted material to the media.

4.   In response to an email stating that the sender had unsuccessfully tried to browse a website that contained the handbook, defendants gave further instructions on how to browse the material.

5.   At least one of the three websites encourages the copying and posting of copies of the allegedly infringing material on other websites. (“Please mirror these files . . . It will be a LOT quicker for you to download the compressed version . . . Needless to say, we need a LOT of mirror sites, as absolutely soon as possible”).

Thus there was associated conduct along with the posting of the addresses of the three websites that led the Judge to conclude that there was contributory infringement. As far as the mere posting of the website addresses was concerned, the Judge found that:

The evidence now before the Court indicates that there is no direct relationship between the defendants and the people who operate the three websites. The defendants did not provide the website operators with the plaintiff’s copyrighted material, nor are the defendants receiving any kind of compensation from them. The only connection between the defendants and those who operate the three websites appears to be the information defendants have posted on their website concerning the infringing sites. Based on this scant evidence, the Court concludes that plaintiff has not shown that defendants contributed to the infringing action of those who operate the infringing websites.

From this case it can be concluded that merely posting a link does not amount to contributory infringement. There must be more active conduct on the part of the person posting the link.

Linking is an essential feature of the world wide web and is vital to its structure. Most websites contain links to others. Links make information on the web manageable and indeed has made internet use accessible to ordinary computer users. Links make it easy to find information available on topics of interest. Thus to restrict linking may have consequences for the effective utilisation of the web that shall be considered in later in this post.

The mere provision of a link (http://www.law.auckland.ac.nz, for example) does not infringe the reproduction right of the linked-to website, because it does not copy the website’s content. There is no copyright in a website address. The linking party merely provides a Universal Resource Locator (URL) that directs the user to the content of the destination webpage.

There is a firm policy basis for this.

It does not make sense to adopt a rule that could lead to the liability of countless parties whose role in the infringement is nothing more than setting up and operating a system that is necessary for functioning of the internet, even where the internet provider has knowledge of potential copyright infringement by its subscribers.[20]

Thus, a linking party is not liable for direct copyright infringement. This was also the case in Perfect 10 v Google[21] where it was held that the operator’s act of framing in-line linked full-size images of copyrighted photographs was not “display” of owner’s works and that

However, there is a difference between mere linking — that is, providing a hypertext linking to another user’s homepage — and other forms of access to material on the web. Most of the cases in this area have focused on copyright law to provide a remedy for what is perceived as an unjustifiable interference with material that has been posted on the web.

5.               The cases

Shetland Times

The case of Shetland Times Ltd v Wills[22] is an example of a problem that arose as a result of deep linking. Although it is a case that is regularly discussed in the context of linking, it does not present a final fully reasoned conclusion after full argument because the case settled before it went to a full substantive hearing. Nevertheless, it is useful in identifying some of the issues that arise.

Shetland Times Ltd was, in Scottish terminology, the pursuer. It reproduced certain news items from their own published newspaper, The Shetland Times, on its website. Access to a news item was gained by clicking on the appropriate headline on the home page of the website. The defender, or defendant, was the managing director of a news reporting service — Shetland News Ltd which ran a news reporting service from its website. On its home page Shetland News included a number of headlines from the pursuer’s home page. These headlines performed the function of hypertext links to the individual news items appearing on the pursuer’s website. A Shetland News user could click on the headline and get direct access to the news item appearing on the pursuer’s website. This access avoided the necessity of going to the pursuer’s home page. There was an objection to this on the part of the pursuer for a commercial reason — the home page was designed to include advertising. If access to news items could be made by circumventing access to the pursuer’s home page, then that home page would lose much of its attraction for potential advertisers and would sustain a loss of advertising revenue.

The pursuer brought the action claiming its headline constituted a cable programme and its inclusion by Shetland News in a cable program service was a primary infringement of copyright under s 20 of the Copyright, Designs and Patents Act 1988 (UK). Because it was an injunction case it was held that a prima facie case had been established for finding, first, that the pursuer’s website on the internet was a cable program service within the meaning of the Act and, secondly, a hypertext link such as the headline contained within that website was a cable programme, again within the meaning of s 7 of the Act. Thirdly, an unauthorised inclusion of a cable programme within a cable program service is an infringement of copyright under s 20.

By inference, therefore, Shetland News’ website also comprised a cable programme service.

The matter only went as far as the interim interdict stage and, as I have said, the issues were not fully canvassed at a full hearing as a settlement was reached. The issues that needed to be resolved were:

1.   whether the internet is a cable programme service under the UK Act; and

2.   an examination of the technological issues including the necessity of copying in order to browse a website and the concept of links to other websites as in themselves attracting copyright protection themselves.

Within the academic community the debate continues over the whether the inclusion of hypertext links to other websites are permissible as being the equivalent of, say, a footnoted reference to another text within traditional published texts.[23]

It has been argued that hypertext links effectively allowing incorporation of another author’s original work within the first work constitute an infringement of copyright but the question of copyright in a hypertext reference itself, either as a literary work or as cable programme has not arisen.

James Connolly and Scott Cameron, in the article “Fair Dealing in Webbed Links of Shetland Yarns”,[24] call for legislation to specifically deal with the issue of copyright and links. They conclude that the technology of the internet is unique and does not fit easily onto existing copyright structure. There is a conflicting ideology on the internet because it originally evolved in universities for the free exchange of information but now has a highly commercialised marketing and sales purpose. These two purposes currently exist side by side. It was Connolly and Cameron’s view that the Shetland Times case was best dealt with as one of fair dealing but if similar questions were to arise again and there was no question of news being involved, it was their opinion that legislation would be needed before any balance between the idealism of the internet and the need to protect intellectual property is found.

Sue French, in a comment on the Shetland Times case,[25] notes that the New Zealand Copyright Act 1994 confers generous powers on the Governor-General to make amending regulations. In recognition of the fact that modern technological developments were overtaking the legislature, the intention of the original drafters of the Copyright Bill was to provide a flexible means by which account could be taken of both technological developments and difficulties of interpretation or application. She said that it is to be hoped that any subsequent judicial finding that would seek to limit or prohibit the use of hypertext links within non-interactive websites would be viewed as an opportunity for appropriate use to be made of the Governor-General’s amending powers.[26]

The Washington Post Company & Ors v Total News Inc & Ors[27]

This was a case about framing that also settled before final determination. However, once again the way the case proceeded and the terms of settlement give an indication of the views of the parties and their legal advisers.

Total News provided a web-based news service in which their site incorporated hyperlinks to websites of the plaintiffs. The links were created in such a way that the plaintiffs’ site did not entirely fill the screen when the link was activated by the user. Rather, part of the site was inserted in a window on Total’s site that occupied only part of a screen. The plaintiffs’ URL did not appear in the address section of the browser, and there was potential for confusion as to the origin of the information in the framed window. From the claim filed by the plaintiffs[28] the major concern was that the actions of Total News interfered with the plaintiff’s contractual arrangements with their advertisers. Total News’ activities offered competing advertising material and devalued the material on the plaintiffs’ sites.

The case settled in June 1997 and the terms of the settlement contain the following provisions:

•    the defendants agreed to permanently cease the practice of “framing” the plaintiffs’ websites as that practice was described in the plaintiffs’ complaint;

•    the defendants were allowed to link to the plaintiffs’ websites only via hyperlinks consisting of the names of the linked sites in plain text, which may be highlighted;

•    the defendants were not allowed to link in any manner reasonably likely to imply affiliation with the plaintiffs, cause confusion or “dilute” the plaintiffs’ trademarks.

As far as New Zealand is concerned there could well be an issue of misleading and deceptive conduct under the Fair Trading Act 1986 if framing causes the browser of a site to believe that there is approval by the owner of the framed site that it (the site) may be displayed, or alternatively, that the absence of the URL leads the browser to believe that the material in the framed site originates from the site being viewed and not another site.

Futuredontics Inc v Applied Anagramic Inc[29]

Although there was a decision in this case involving framing, the issue is not entirely resolved. Applied Anagramic was a dental services website. It framed content from a competing site. The frames included information about Applied Anagramic, its trademark and links to its webpages.

The Court held that a website containing a link that reproduced webpages within a frame may amount to an infringing derivative work. In addition, the modifications incorporated in the frame could, without authorisation, amount to infringement.

The Court’s approach seems to be directed towards the way in which the link functioned to incorporate content, and the decisions seems directed more to the use of the material than to the link itself in that the content was incorporated as part of the ‘get up” of the site.

Ticketmaster Corp v Microsoft[30]

Ticketmaster Corp v Microsoft is another deep linking case that settled before trial. Once again, it is of interest in that it highlights some of the issues behind linking.

Microsoft operated a website known as “Seattle Sidewalks”. It provided deep links to Ticketmaster’s website, thus avoiding the initial pages of the Ticketmaster site that contained advertisements, products and services and enabled users of the Sidewalks site to engage in the online purchase of tickets from Ticketmaster.

Ticketmaster claimed that Microsoft had wrongfully appropriated and misused its name and trademarks, and engaged in unfair competition. Ticketmaster did not attack linking as such, but Microsoft’s use of deep hyperlinking, which diminished the advertising revenue available to it and increased Microsoft’s own advertising revenue.

The case was closely watched by internet experts who hoped for clear and considered judicial guidance on the question of linking. The settlement in February 1999 resulted in an agreement by Microsoft not to provide deep links to Ticketmaster’s site but to link to Ticketmaster’s homepage. Following this, Ticketmaster has allowed deep linking by sites such as Yahoo!, but only after the parties had entered into a linking agreement.

Ticketmaster Corp v Tickets.com[31]

In this case Tickets.com was in competition with Ticketmaster. Once again, there was deep linking to Ticketmaster’s site. In addition, Tickets.com copied material from the Ticketmaster site and posted false information about the availability of tickets from Ticketmaster.

Ticketmaster alleged copyright infringement, breach of contract and unfair competition.

In considering (and dismissing) the claim of copyright infringement based on hyperlinking, Judge Harry Hupp said:

“Hyperlinking does not itself involve a violation of the Copyright Act (whatever it may do for other claims) since no copying is involved. The customer is automatically transferred to a particular genuine webpage of the original author. There is no deception in what is happening. This is analogous to using a library’s card index to get reference to particular items, albeit faster and more efficiently.”[32]

Ticketmaster’s claim for breach of contract was also dismissed. Ticketmaster had posted terms and conditions on its website prohibiting deep linking, but the Judge held that Ticketmaster’s complaint did not provide a factual foundation that Tickets.com knew of or agreed to these terms.

The Judge dealt with the claim of unfair competition by deep linking in this way:

The complaint also alleges deep linking as an example of unfair competition, but the court concludes that deep linking by itself (i.e. without confusion of source) does not necessarily involve unfair competition.

It is suggested that this case does not definitively decide when it is permissible to deep link and when it is not. It is a starting point, although it was a short judgment on an interlocutory application without full evidence or argument on the substantive issues.

There are also issues left open regarding copyright infringement, unfair competition and misleading or deceptive conduct where there is confusion as to the source of the content, or ignorance as to the fact that there has in fact been a link to a separate and distinct site and deep within its interior.

6                 Linking, anti-circumvention and copyright

6.1             Universal City Studios v Reimerdes & Corley

The case of Universal City Studios v Reimerdes and Corley[33] is primarily about the anti-circumvention provisions of the American Digital Millennium Copyright Act. This discussion focuses on the way in which the Court approached a hypertext link and characterised it as the functional equivalent of trafficking in anti-circumvention devices. To extend this reasoning to other linking activity could have significance for internet utility and it is for this reason that a detailed study of this aspect of the case is warranted.

6.1.1         Background

The plaintiff movie companies owned the copyrights in a large number of films. Many of these films were (and are) available on DVD. Films on DVD disks are in digital format. Without some form of copy protection the files could easily be copied and distributed thus infringing the plaintiffs’ copyright and causing them loss of revenue. The plaintiffs applied an encryption to their DVDs known as CSS. This encryption can be “unlocked” by DVD players so that the movie on the disk can be viewed. It prevents the copying of the movie files from the DVD.

Jon Johansen, a Norwegian student, reverse engineered the CSS encryption and developed a utility known as DeCSS. This unlocks the CSS encryption on a DVD disk and allows the files to be copied to a computer hard disk drive. Johansen posted the DeCSS code on the internet, thus making the program available to the world.

In November 1999 the defendants posted DeCSS on their website and made it available for download. They established links from their webpage to other sites where DeCSS could be similarly obtained.

The movie companies became aware of this and late in 1999 sent “cease and desist” letters to a number of site operators, some of whom complied. The defendants did not. In January 2000 proceedings were commenced against them. The plaintiffs obtained preliminary injunctive relief prohibiting the defendants from making DeCSS available for download from their site. The defendants, in an act of electronic civil disobedience, provided links to other sites offering DeCSS

The plaintiffs relied on the provisions of the Digital Millennium Copyright Act, which was enacted by the US Congress in 1998.

6.2             The categorisation of linking activity: the decision in Universal City Studios v Reimerdes

Section 1201(a)(2) of the Copyright Act, part of the DMCA, provides that:

No person shall . . . offer to the public, provide or otherwise traffic in any technology . . . that —

(A)   is primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access to a work protected under [the Copyright Act];

(B)    has only limited commercially significant purpose or use other than to circumvent a technological measure that effectively controls access to a work protected under [the Copyright Act]; or

(C)    is marketed by that person or another acting in concert with that person with that person’s knowledge for use in circumventing a technological measure that effectively controls access to a work protected under [the Copyright Act].

The italicised words are the subject of expansive interpretation in the context of the linking activities of the defendants.

There is no doubt whatsoever that DeCSS was designed and produced for the purpose of circumventing the technological measure (CSS encryption on DVDs) that controls access to the files on the DVD. By posting DeCSS on its website for downloading the defendants obviously breached s 1201(a)(2)(A) in that they offered it to the public and provided it.

In his consideration of linking Judge Kaplan dealt first with an interpretation of the Act and then whether or not linking may constitute protected speech pursuant to the First Amendment to the US Constitution.

Interpretation

The Judge’s reasoning on interpretation is as follows:[34]

“Linking to Sites Offering DeCSS

Plaintiffs seek also to enjoin defendants from “linking” their 2600.com website to other sites that make DeCSS available to users. Their request obviously stems in no small part from what defendants themselves have termed their act of “electronic civil disobedience”— their attempt to defeat the purpose of the preliminary injunction by (a) offering the practical equivalent of making DeCSS available on their own website by electronically linking users to other sites still offering DeCSS, and (b) encouraging other sites that had not been enjoined to offer the program. The dispositive question is whether linking to another website containing DeCSS constitutes “offer[ing DeCSS] to the public” or “provid[ing] or otherwise traffic[king]” in it within the meaning of the DMCA. Answering this question requires careful consideration of the nature and types of linking.

Most webpages are written in computer languages, chiefly HTML, which allow the programmer to prescribe the appearance of the webpage on the computer screen and, in addition, to instruct the computer to perform an operation if the cursor is placed over a particular point on the screen and the mouse then clicked. Programming a particular point on a screen to transfer the user to another webpage when the point, referred to as a hyperlink, is clicked is called linking. Webpages can be designed to link to other webpages on the same site or to webpages maintained by different sites.”

Judge Kaplan then moved on to define three categories of links on websites as:

•    websites that contain information of various types, but no link to DeCSS but that contain links to other sites containing DeCSS that the user must follow to obtain the program;

•    links that take the user to another site where there is information in addition to a direct link to the DeCSS program;  and

•    links to sites where DeCSS may be downloaded by merely clicking on the link.

He then proceeded to consider whether the links provided by the defendants were within the ambit of the statute.

“To the extent that defendants have linked to sites that automatically commence the process of downloading DeCSS upon a user being transferred by defendants’ hyperlinks, there can be no serious question. Defendants are engaged in the functional equivalent of transferring the DeCSS code to the user themselves.”[35]

Here Judge Kaplan is referring to the third category and develops the suggestion that such a link is the “functional equivalent” of making the program available on the originating link site. This overlooks the fact that the site linked to may be outside the jurisdiction or in any respect beyond the control of the administrator of the originating site.

“Substantially the same is true of defendants’ hyperlinks to webpages that display nothing more than the DeCSS code or present the user only with the choice of commencing a download of DeCSS and no other content. The only distinction is that the entity extending to the user the option of downloading the program is the transferee site rather than defendants, a distinction without a difference.”

Judge Kaplan refers to the second category that provides a link to a site where DeCSS may be downloaded, but that requires a positive act (clicking on the download icon) to obtain the DeCSS program. The Judge expands the concept of functional equivalence beyond an automatic download to an action that must be initiated by a third party.

“Potentially more troublesome might be links to pages that offer a good deal of content other than DeCSS but that offer a hyperlink for downloading, or transferring to a page for downloading, DeCSS. If one assumed, for the purposes of argument, that the Los Angeles Times website somewhere contained the DeCSS code, it would be wrong to say that anyone who linked to the Los Angeles Times website, regardless of purpose or the manner in which the link was described, thereby offered, provided or otherwise trafficked in DeCSS merely because DeCSS happened to be available on a site to which one linked. But that is not this case. Defendants urged others to post DeCSS in an effort to disseminate DeCSS and to inform defendants that they were doing so. Defendants then linked their site to those “mirror” sites, after first checking to ensure that the mirror sites in fact were posting DeCSS or something that looked like it, and proclaimed on their own site that DeCSS could be had by clicking on the hyperlinks on defendants’ site. By doing so, they offered, provided or otherwise trafficked in DeCSS, and they continue to do so to this day.”

Thus the Judge acknowledges a problem because DeCSS is not automatically available or one user-initiated step distant from such availability. “Functional equivalence” is not available. A motive based test is created, whereby purely information-giving sites like the Los Angeles Times are differentiated from sites that make information available with other than a “pure” motive. There can be no doubt that this raises free speech and freedom of information issues, along with what seems to be an overly strained approach to the reality of the technology of the world wide web in an effort to justify the result reached.

There are two areas that need to be examined in testing the validity of the reasoning employed. The first uses a real world analogy. The second considers whether the legislation can be interpreted as expansively as it has been in this case.

The real-world analogy

Judge Kaplan’s categories bear examination within the context of the real world. This section works backwards and takes category 3 as the first example.

A real-world equivalent to the scenario advanced is where 2600.com occupies a storefront location and has a sign pointing to another storefront indicating that DeCSS can be obtained from the other storefront. In the other store is a bin where a party may uplift DeCSS free.

Category 2 continues the scenario. There is the same sign pointing to the same destination storefront. However, rather than uplifting a copy of DeCSS from a bin, the party seeking the program must make a specific inquiry to the operator of the storefront. Upon such a request being made, the operator of the other storefront makes a copy of DeCSS available.

Category 1 takes the scenario further. The person seeking DeCSS goes to the 2600 storefront and is directed by a sign to another storefront. At that other storefront are books and magazines about DeCSS but not copies of the program. Instead, there is another sign pointing to a third storefront. At that third storefront DeCSS is available in both free bins and upon request from the third storefront proprietor.

A common feature of all three scenarios is that the 2600 storefront has provided information and nothing more. The information that it has given has facilitated the location of DeCSS by a person interested in obtaining that program. If the 2600 storefront made DeCSS available that would be an infringement. However, directing a person to a place where the program can be uplifted cannot be considered the same as making it available at the DeCSS counter. When analysed in this way it becomes clear that a differing measure results for behaviour in the online environment and the real-world one.

Extending the scope

The Judge’s definition of “to traffic” is from the 1971 edition of the Compact Edition of the Oxford English Dictionary. “Trafficking” is a word that has a connotation of either commercial activity or supply of a product (normally of contraband). Further, the word suggests that the person “trafficking” does something positive in the chain of supply, rather than passively provide information. For example, a person who says that cannabis can be obtained from a particular address cannot be charged with dealing or supplying cannabis. Indeed, the Judge himself points out that “engag(ing) in dealing” is an aspect of trafficking. This implies active participation in the provision of the contraband. As far as that word is concerned, the giving of information about a site where DeCSS is available cannot be trafficking, in that providing a link is a form of giving information and directing the user to the source of that information.

The definition of “provide” as “make available” or “furnish” is unremarkable. However, once again the word carries a requirement of active involvement in the process of transfer of a product. It is drawing a long bow to say that by giving information about the location of a product one is engaging in providing the product.

The Judge then moves to the words “otherwise traffic in” as modifiers of the words “traffic” and “provide”. He says that the true definition means that one who presents, holds out or makes DeCSS available, knowing its nature and for the purpose of allowing others to acquire it is in breach.

Then the concept of “functional equivalence” is introduced. By providing links to sites where DeCSS is available or by which DeCSS may be obtained, the defendants were engaged in the “functional equivalent” of transferring the code to the user themselves. The terms “functional equivalent” makes its first appearance in the discussion but is not in any way defined. Symbolism, expression and functionality are terms that are well known in copyright law as aspects of the “idea/expression dichotomy”. It seems that the Judge has used the phrase as a convenient way of making the defendants liable where on a rigorous interpretation of the statute such liability does not exist.

Links — Kaplan J’s approach

A link provides information about where the programs may be located. This is rather like a phonebook address or a footnote in that it directs the user to a location. It is the webmaster of that location who actively engages in the act of supply. The use of the term “functional equivalent” suggests that by providing a link to another site, the link provider is a provider of the software. That defies reality and the clear interpretation of the language of the statute that neither directly nor by implication allows the insertion or utilisation of the concept of functional equivalence.

Links to 2600 and its mirror as well as a number of other sites that could lead to DeCSS were available via no less an organ than the New York Times online in an article in the Cyberlaw Journal for 16 June 2000 entitled “Is linking illegal?”.[36]

The Judge would exempt the New York Times from the functional equivalent of transferring the code by providing such links for the reasons associated with his discussion of the availability of DeCSS via the Los Angeles Times. His sole justification for differentiating the actions of the defendants from those of the Los Angeles Times and New York Times was a matter of motive. Clearly he was incensed by what the defendants described as “electronic civil disobedience” and was prepared to extend the language of the statute to apply what was a punitive application of the remedy of injunctive relief when the appropriate remedy may have been a consideration of contempt of Court. It is doubtful, however, that such remedy may have been applicable.

There can be no difference between what the New York Times did and what the defendants did. They both provided information. The effect is that those using the links of either source may be provided with DeCSS. The issue of motive is irrelevant. DeCSS may still be obtained. If the Judge were to be consistent, the New York Times is as culpable as the defendants.

First Amendment, jurisdictional and technical issues[37]

Judge Kaplan’s consideration of the applicability of the First Amendment to the defendant’s linking was as follows.[38]

Judge Kaplan considered the applicability of the First Amendment to linking. He started by observing that the DMCA applies to links deliberately created by a website operator for the purpose of disseminating technology that enables circumvention of access controls on copyrighted works.

He likened links to road signs but observed that links are more functional in that unlike road signs they take one almost instantaneously to the desired destination by the click of a mouse.

His Honour observed that like computer code in general, links have both expressive and functional elements and are within the area of First Amendment concern.

He observed that the application of the DMCA to the type of linking in the case would serve the same government interest as its application to the posting by the defendants of the DeCSS code. He further observed that posting of content and linking amount to very much the same thing and thus the regulation of the linking at issue was unrelated to the suppression of free expression.

Judge Kaplan stated that preventing the defendants from linking to a site in respect of which the plaintiffs could take legal action in any event would advance the statutory purpose of preventing dissemination of circumvention technology.

He noted that links are what unify the world wide web into a single body of knowledge and what makes the web unique; that links are the mainstay of the internet and indispensable to the convenient access to the vast world of information. He observed that they are used in ways that do a great deal that promote the free exchange of ideas and information – a central value of US culture. He noted that anything that would impose strict liability on a website operator for the entire contents of any website to which the operator linked could cause grave constitutional concerns. Website operators could be inhibited from linking for fear of exposure to liability. It was equally clear that exposing those who use links to liability under the DMCA might inhibit their use, as some website operators, confronted with claims that they may have posted circumvention technology falling within the statute, may be more inclined to remove the allegedly offending link rather than test the issue in Court.

Thus, Judge Kaplan viewed with concern the negative effect of a rule permitting liability for or injunctions against internet hyperlinks. He stated, however, that that was not unique to linking. The law relating to defamation provided an analogy. The threat of defamation suits creates the same risk of self-censorship, the same effect for the traditional press as a prohibition of linking to sites containing circumvention technology poses for website operators. In the same way that defamation does not immunise the press from all actions for defamation, DMCA liability cannot utterly immunise website operators from all actions for disseminating circumvention technology. The solution for the problem he decided was the same — the adoption of a standard of culpability sufficiently high to immunise the activity whether it is the publishing of a newspaper or linking, except in cases in which the conduct in question has little or no redeeming constitutional value.

In defamation he pointed out that there is a two-tiered constitutional standard. There may be no liability under the First Amendment for defamation of a public official or a public figure unless a plaintiff proves by clear and convincing evidence that the defendant published the offending statement with knowledge of its falsity or with serious doubt as to its truth. Liability in private figure cases, on the one hand, may not be imposed in the absence of proof at least of negligence. He considered that a similar approach would minimise any detrimental effect here.

Another argument that was advanced was that liability based on linking to another site simply because it contained circumvention technology could be overkill but the Judge considered that the offence under the DMCA had an essential ingredient that was a desire to bring about the dissemination, thus a strong requirement of that forbidden purpose is an essential prerequisite to liability for linking.

Thus Judge Kaplan held that there could be no injunction against, nor liability for linking to a site containing circumvention technology — the offering of which is unlawful under the DMCA — without clear and convincing evidence that those responsible for the link:

(a) know at the relevant time that the offending material is on the link-to site;

(b)    know that it is a circumvention technology that may not lawfully be offered; and

(c) create or maintain the link for the purpose of disseminating that technology.

The Judge found that the plaintiffs had established by clear and convincing evidence that these criteria had been fulfilled and that the anti-linking injunction on the facts did no violence to the First Amendment.

There are five issues that arise out of this section of the decision.

1. Content-based or content-neutral? The first issue relates to a point that arises earlier in a consideration of whether DeCSS is entitled to First Amendment protection. In essence, the Judge distinguished between content-based speech (which will receive strict scrutiny and protection — the issue being the protection of what is being said) and content-neutral speech (where restrictions have nothing to do with expression and a less exacting standard is applied because the restrictions are not motivated by a desire to limit the message). He considered that the application of the Digital Millennium Copyright Act (DMCA) was a content-neutral form of regulation. He found that DeCSS has a functional non-speech aspect which is to decrypt CSS files. He found that the anti-trafficking provisions of the DMCA had nothing to do with restricting the ideas of programmers, but everything to do with functionality. The provisions also furthered an important governmental interest in protected copyrighted works from the expanded risk of piracy in the digital age.

There may well be subtleties in First Amendment jurisprudence that effectively challenge these conclusions but it appears that the Judge, in considering the tensions that arise between First Amendment speech protections and the interests of copyright holders, is construing more heavily in favour of the importance of copyright protection over the importance of free expression. In addition, in his consideration he is implicitly expressing unjustified fears of a digital future which shall be dealt with at a later stage.

2. Navigation on the web. The second issue is the recognition of the importance of links as a means of navigation around the world wide web and the utility of links that take the user straight to the information sought. Much of what the Judge says in this regard is related to his findings as to the applicability of the First Amendment to DeCSS. He uses the same approach to links as he does to DeCSS in finding that they are not forms of content-based expression but are in fact content-neutral, linking as they do to sites where DeCSS (content-neutral expression) may be downloaded. This means that a less rigorous test is applied, and therein lays a serious problem if this case has precedent value or is upheld on appeal.

Simply put, links will not be entitled to the strict scrutiny available to content-based speech, and judicial regulation of linking on the internet will be substantially increased in all areas — not just that of copyright protection. Although the Judge is dealing with a copyright case and the DMCA, his findings on the applicability of the First Amendment to links cannot be so narrowly construed. Thus, the potential negative effect on linking on the internet is heightened.

In his consideration of linking the Judge applies an interesting test that seems to be based on the intention of the linker. Rather than consider a link as a means of conveying information and connecting the user to the source, the Judge views the motive for the link as determinative of whether or not the linker is entitled to First Amendment protection. In a piece of legislation that does not, in the sections referred to in the decision, contain a concept of principals and parties to breaches of the legislation, the Judge is engrafting a concept of party to a breach by developing a consideration of motive for linking. He acknowledges that there may be no injunction nor liability for linking to a DeCSS site unless there is clear and convincing evidence that those responsible for the link:

•    know at the relevant time that the offending material is on the linked-to site,

•    know that it is circumvention technology that may not lawfully be offered, and

•    create or maintain the link for the purpose of disseminating that technology.

Thus he says that links may be protected unless the motive is impure. He has cast his criteria within the context of the DMCA, but it is quite clear that other criteria can be applied to specific cases as and when they arise.

3. The content neutrality of links? The third issue arises as part of the second. The world wide web is synonymous with the access to information that the internet provides and is built on links as a fundamental part of its architecture. Any rule that inhibits linking could retard the use and development of the web. Can links be considered as purely content-neutral? Or should they be considered as signposts to further information sources and as such carry within them an information-bearing or content-based element that would entitle them to First Amendment protection? Admittedly links contain within them a functional aspect, but that should not be allowed to obscure other aspects inherent within them.[39]

Without referring specifically to the cases, the approach of the 3rd Circuit Court of Appeals[40] and the Supreme Court[41] in Reno v ACLU was to accord to aspects of the internet protection based on the nature of the continuing dialogue that pervades this essentially communicative environment. Part of the facilitation of this dialogue is links. In fact the rise of the world wide web has made the http:// protocol as synonymous with the internet as TCP\IP and is as significant in the development of internet communication as, for example, the mechanisation of Gutenberg’s printing press.  Thus to relegate the link to the level of content-neutral speech underestimates its significance in the process of communication of ideas. If one accords a strict scrutiny test to links, as one must in considering content-based speech, any interference with links must be an interference with a significant aspect of free speech on the internet.

4. Extraterritorial effect of the decision – The real significance of an anti-linking injunction would not be with US websites subject to the DMCA, but with foreign sites that arguably are not subject to it and not subject to suit here. An anti-linking injunction to that extent would have a significant impact and thus materially advance a substantial governmental purpose.”[42]

In making this comment the Judge is recognising the extraterritorial effect that his decision will have on the ability to link to foreign websites. In the borderless world of the internet one may become increasingly concerned at decisions of domestic Courts that may have an impact on perfectly legitimate acts in other jurisdictions.[43] Although there is ample authority to the effect that the internet functions in the real world and should not be ruled by some sort of “virtual governance”, nevertheless, domestic Courts must be aware that in many cases involving this world-wide, international and borderless medium, the ramifications of a decision may extend far beyond domestic borders and have a significant effect on those within other jurisdictions. In a world where imperialism is a now derogatory term, one would be concerned to see the rise of a “virtual” or “electronic” imperialism where one country, perhaps unintentionally, extended its rules and legal values beyond its territory.

In the context of the internet, an injunction restraining the publication of links ceases to have local application, unless some form of technical device can be provided that prohibits access to the linked sites from geographically based IP numbers, as in the case of the Havana House Cigars weblink, where access to the New Zealand-based site was disabled for those attempting to link from New Zealand, based on concerns about breaches of laws relating to advertisement of tobacco products.[44] The site, however, was accessible to those from countries other than New Zealand.[45]

5. Value judgments and technological misunderstandings. The final issue that arises out of the decision in Universal City Studios v Reimerdes and Corley[46] is referred to in the Judge’s introductory remarks and essentially sets the stage for what follows in his consideration of the legislation and the actions of the defendants:[47]

“In an era in which the transmission of computer viruses — which, like DeCSS, are simply computer code and thus to some degree expressive — can disable systems upon which the nation depends and in which other computer code also is capable of inflicting other harm, society must be able to regulate the use and dissemination of code in appropriate circumstances. The Constitution, after all, is a framework for building a just and democratic society. It is not a suicide pact.”

Throughout the decision there is a negative spin placed on the digital environment and the danger to established order (particularly in the intellectual property field) that this environment presents. The starting point with an expression of concern about computer viruses is well founded. The statement that viruses consist of computer code is factually correct. The equating of computer viruses with DeCSS, a decryption program, is a long leap and one, with respect, that cannot be sustained. DeCSS does not disable computer systems upon which the nation depends. Comments about the importance of computer systems appear at a later stage in Judge Kaplan’s decision:[48]

“Society increasingly depends upon technological means of controlling access to digital files and systems, whether they are military computers, bank records, academic records, copyrighted works or something else entirely. There are far too many who, given any opportunity, will bypass those security measures, some for the sheer joy of doing it, some for innocuous reasons, and others for more malevolent purposes. Given the virtually instantaneous and worldwide dissemination widely available via the Internet, the only rational assumption is that once a computer program capable of bypassing such an access control system is disseminated, it will be used.”

Perhaps the Judge was excessively concerned with the activities of Corley and others involved in the past with 2600.com. His fear of the malevolent hacker is obvious and well founded.[49] However, he refers to those who attempt to access systems for malevolent purposes. Those who devised DeCSS were not so destructively minded. Perhaps he was influenced by his perceptions of the activities of Corley as part of his motive-based approach to the posting on links. He describes them in the following way:[50]

“Defendant Eric Corley is viewed as a leader of the computer hacker community and goes by the name Emmanuel Goldstein, after the leader of the underground in George Orwell’s classic, 1984. He and his company, defendant 2600 Enterprises, Inc, together publish a magazine called 2600: The Hacker Quarterly, which Corley founded in 1984, and which is something of a bible to the hacker community. The name “2600” was derived from the fact that hackers in the 1960s found that the transmission of a 2600 hertz tone over a long distance trunk connection gained access to “operator mode” and allowed the user to explore aspects of the telephone system that were not otherwise accessible. Mr Corley chose the name because he regarded it as a “mystical thing,” commemorating something that he evidently admired. Not surprisingly, 2600: The Hacker Quarterly has included articles on such topics as how to steal an internet domain name, access other people’s email, intercept cellular phone calls, and break into the computer systems at Costco stores and Federal Express. One issue contains a guide to the federal criminal justice system for readers charged with computer hacking. In addition, defendants operate a website located at <http://www.2600.com&gt; (“2600.com”), which is managed primarily by Mr Corley and has been in existence since 1995.

Prior to January 2000, when this action was commenced, defendants posted the source and object code for DeCSS on the 2600.com website, from which they could be downloaded easily. At that time, 2600.com contained also a list of links to other websites purporting to post DeCSS.”

The Judge went on to describe some of Corley’s actions after litigation commenced. He noted the defendants did not present affidavits of evidence at the hearing and that the Court granted a preliminary injunction to prevent the defendants from posting DeCSS. The defendants then removed DeCSS from their website but in the name of “electronic civil disobedience” they did not remove links to other websites that allegedly offered DeCSS. The list of these websites grew to nearly 500 by July 2000 and with reference to the lawsuit carried the messages “Stop the MPAA” and “We have to face the possibility that we could be forced into submission. For that reason it’s especially important that as many of you as possible, all throughout the world, take a stand and mirror these files.”[51]

The Judge noted that the defendants were trying to hinder the plaintiff’s case as much as possible as some of the links on the mirror list allowed the user to download DeCSS and successfully decrypt a CSS-encrypted DVD.

It seems that the Judge allowed his perception of Corley and defendants as hackers bent on the invasion and destruction of computer systems to be transferred to the particular (and less threatening) activity in this case. There can be no similarity between DeCSS and a system that allows one to access the Pentagon other than that they may share the same programming language. It appears the Court was offended by the use of links as a means of circumventing the Court order and was determined to deal with the defendants in a punitive way by an expansive interpretation of the legislation, and the introduction of the functional equivalence test.

6.3             Universal City Studios v Reimerdes and Corley — the appeal

Not surprisingly there was an appeal against Judge Kaplan’s decision. The appellant was represented by Kathleen Sullivan, Dean of Stanford Law School and amici briefs were filed by a number of organisations.

The appellant’s approach was grounded on a First Amendment argument rather than the issue of challenging Kaplan J’s approach to “functional equivalence” or a consideration of the violence done to traditional copyright theory posed by the DMCA. Although the decision as a whole is an interesting one, this discussion is limited to the issue of linking.

Linking

Judge Kaplan’s prohibition of linking to other sites, where DeCSS was available, was considered. The Court considered the nature of hyperlinks and observed that Judge Kaplan had identified speech and non-speech components in hyperlinks. The Court focused on an apparent difference in expression and functionality. The act of linking utilised functionality rather than expression, and that can be done without reference to content, thus rendering the functional aspect of link content neutral. Applying the O’Brien/Ward/Turner[52] analysis for content-neutral regulation, the prohibition served a substantial government interest unrelated to the freedom of expression. The Court observed that a three-step knowledge based test about the effect of the link had to be applied, which required:

Clear and convincing evidence that those responsible for the link:

(a)  know at the relevant time that the offending material is on the linked-to site;

(b) know that it is circumvention technology that may not lawfully be offered; and

(c)  create or maintain the link for the purpose of disseminating that technology.

Although the Court found it unnecessary to consider the test as developed, it rejected the appellant’s contention on the basis that:

[The appellants] had ignored the reality of the functional capacity of decryption computer code and hyperlinks to facilitate instantaneous unauthorized access to copyrighted materials by anyone anywhere in the world. Under the circumstances amply shown by the record, the injunction’s linking prohibition validly regulates the Appellants’ opportunity instantly to enable anyone anywhere to gain unauthorized access to copyrighted movies on DVDs.[53]

Indeed, there is a further irony that arose from the Universal City Studios case itself. In accordance with the injunction, Corley removed the hypertext links from the 2600 website. However, the text of those links remained.  Those who wished to access the sites merely needed to transcribe the URL into the File\Open dialogue of a browser and the site could be reached. Thus the information remains without the actual linking function. This emphasises the fallacy of considering a link as the functional equivalent of direct access from the 2600 site. Indeed, it is likely that the Court could run into serious First Amendment issues if it tried to enjoin Corley from publishing the text of the links. Thus, this emphasises the care that must be adopted in addressing novel technological concepts.


[1]        David Harvey internet.co.nz  2nd ed (LexisNexis, Wellington 2005)

[2]        [2011] SCC 47

[3]           487 F.3d 701 (2007)

[4]           Nils Svensson, Sten Sjögren, Madelaine Sahlman, Pia Gadd v Retreiver Sverige AB http://curia.europa.eu/juris/document/document.jsf?text=&docid=130286&pageIndex=0&doclang=EN&mode=lst&dir=&occ=first&part=1&cid=1258343 (last accessed 19 March 2013). For the submission see http://www.ivir.nl/news/European_Copyright_Society_Opinion_on_Svensson.pdf (last accessed 19 March 2013).

[5]           [2000] WL 1160678 SDNY 2000 (17 August 2000), 111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001)

[6]        The issue of linking and framing was considered by Rick Shera in 1998  in a post-graduate thesis entitled Linking, Framing and Copying – A New Zealand Perspective.  http://www.lojo.co.nz/news/Internet-Linking-and-copyright-A-blast-from-the-past?i=109; http://www.lojo.co.nz/downloads/0070525001361505661.pdf It would be crass and churlish of me not to acknowledge Ricks early ground-breaking contribution to the scholarship in this field

[7]        Geoff McLay, Christopher Murray, Jonathan Orpin New Zealand Law Style Guide (Thomson Reuters, Wellington 2009)

[8]        [2000] WL 1160678 SDNY 2000 (17 August 2000), 111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001) .p 46. Emphases added.

[9]        487 F.3d 701 (2007)

[10]      [2011] SCC 47

[11]      H C Auckland, 19 January 1998, CP 344/97 Master Kennedy-Grant

[12]      [2013] EWCA Civ 68

[13]      [2012] NZHC 2352

[14]      [2013] FCA 60

[15]      [2012] VSC 533

[16]      For the position in Australia see Timothy Hughes, “Intellectual Property and Browsing the Web”, in Internet Law Anthology, Sydney, Prospect Media Pty Ltd, 1997) at 55. For the position in the US see Gregory M Luck, “The Implied Licence: An Evolving Defence to Patent Infringement” (1998) IP Litigator US 13.

[17]      No 2: 99-CV-808C US District Court, D Utah, Central Division, 6 December 1999 http://eon.law.harvard.edu/property00/metatags/ULM.html (last accessed 21 March 2013) emphasis added.

[18]      Contributory infringement has developed as a part of US copyright theory. Assuming there is an act of direct infringement, contributory infringement may be found when “[o]ne who, with knowledge of the infringing activity, induces, causes or materially contributes to the infringing conduct of another.” Fonovisa, Inc. v. Cherry Auction, Inc., 76 F.3d 259, 264 (9th Cir. 1996) (citing Gershwin Publishing Corp. v. Columbia Artists Mgmt., Inc., 443 F.2d 1159, 1162 (2d Cir. 1971)). See generally Newborn v. Yahoo!, Inc., 391 F.Supp.2d 181 (D.D.C. 2005); Monotype Imaging, Inc. v. Bitstream, Inc., 376 F.Supp.2d 877, 883 (N.D.Ill. 2005).

The knowledge requirement is objective and is satisfied where the defendant knows or has reason to know of the infringing activity. See, e.g., Sega Enters. Ltd. v. MAPHIA, 948 F. Supp. 923, 933 (N.D. Cal. 1996); Casella v. Morris, 820 F.2d 362, 365 (11th Cir. 1987). However, if a defendant can show that its product was capable of substantial or commercially significant non-infringing uses, then constructive knowledge of the infringement should not be imputed. Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417 (1984).

With respect to the materiality requirement, the Ninth Circuit has held that providing the site and facilities for known infringing activity is sufficient. Fonovisa, 76 F.3d at 264 (adopting the analysis of Columbia Pictures Indus., Inc. v. Aveco, Inc., 800 F.2d 59 (3rd Cir. 1986)); see also 3 Nimmer § 12.04[A][2][b]. However, some courts have required the participation to be “substantial,” finding the mere fact that equipment or facilities may be used for copyright infringement is not determinative. Apple Computer, Inc. v. Microsoft Corp., 821 F. Supp. 616, 625 (N.D. Cal. 1993), aff’d, 35 F.3d 1435 (9th Cir. 1994); see also Sega Enters. Ltd. v. MAPHIA, 948 F. Supp. at 933; Religious Tech. Ctr. v. Netcom On-line Communications Servs., Inc., 907 F. Supp. 1361 (N.D. Cal. 1995); Perfect 10 v. Visa Int’l Service Ass’n, 2004 WL 1773349, at *4 (N.D.Cal. Aug.5, 2004) (must show “a relationship between the ··· services provided by the [d]efendant[ ] and the alleged infringing activity as opposed to the mere operation of the website businesses.”).

[19]      This factual scenario bears a certain similarity to the actions of Corley in Universal City Studios v Reimerdes 111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001) (Appeal Court Decision F Supp 2d 346) – for discussion see below

[20]      See Religious Tech Ctr v Netcom On-line Commun Servs, Inc, 907 F Supp 1361, 1370, 1372 (ND Cal 1995).

[21]      487 F.3d 701 (2007) 9th Cir. For discussion see below.

[22]      (1997) EMLR 277, (1996) 37 IPR 71.

[23]      In the content of a link as a footnote see Crookes v Newton [2011] SCC 47, [2011] 3 S.C.R. 269 – for discussion  of this case see below.

[24]      James Connolly and Scott Cameron, “Fair Dealing in Webbed Links of Shetland Yarns” 1998, 2 JILT: http://www2.warwick.ac.uk/fac/soc/law/elj/jilt/1998_2/connolly/ (last accessed 21 March 2013).

[25]      [1998] NZLJ 41.

[26]      Further analysis of linking is to be found in Cavazos and Miles “Copyright on the WWW; Linking and Liability” (1997) 4 Richmond Journal of Law & Technology http://law.richmond.edu/jolt/v4i2/cavazos.html (last accessed 21 March 2013). As it turned out it was not necessary for the Governor-General to exercise such power and there were amendments to the Copyright Act in 2008 which were the subject of the Copyright (New Technologies) Amendment Act 2008.

[27]      97 Civ 1190 (PKL).

[28]      For details see John Lambrick “Protecting Content in an On-Line Environment” Telecommunication Jnl of Australia Vol 58, Issue 4, p 53; and John Lambrick “Hyperlinking, Framing and Copyright — Waiting for the Dust to Settle” 12 Computers & Law 11 at p 12.

[29]      1997 46 USPQ 2d 2005 (CD Calif 1997) 1998 US Dist LEXIS 2265 (CD Cal 1/30/98). Upheld 1998 WL 417413, unpub’d opinion (9th Cir 7/23/98): http://eon.law.harvard.edu/property99/metatags/1998futu.html (last accessed 21 March 2013).

[30]      Case No 97-3055 DDP (CD Cal, complaint filed, 28 April 1997; amended complaint filed February 1998).

[31]      2000 US Dist LEXIS 4553

[32]      Of course there is a nonsense in what the Judge says. The customer isn’t transferred anywhere. The user remains static. Information is downloaded to a user’s computer. Apart from that technological error the Judge’s comment about linking and reference systems is echoes in Crookes v Newton [2011] SCC 47

[33]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001).

[34]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001), p 46.

[35]      The emphasis is mine.

[36]      Carl Kaplan, “Is linking illegal?” Cyber Law Journal, 16 June 2000: http://www.nytimes.com/library/tech/00/06/cyber/cyberlaw/16law.html (last accessed 21 March 2013).

[37]      Although the First Amendment is a “local ordinance” and the discussion is in terms of the US jurisprudence on the subject of freedom of expression, the developing interest of commentators in the applicability of freedom of speech in the context of copyright makes this discussion relevant.

[38]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001), p 76.

[39]      For further discussion see Crookes v Newton [2011] SCC 47

[40]      929 F Supp 824 (1996).

[41]      521 US 874 (1997).

[42]      Universal City Studios v Reimerdes 111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001). However, a generous view of extraterritorial jurisdiction in many civil and some criminal cases may challenge this comment.

[43]      See LICRA v Yahoo! Order in Summary Proceedings by the Superior Court of Paris rendered on 22 May 2000 by First Deputy Chief Justice Judge Jean-Jacques Gomez http://www.lapres.net/yahweb.html (last accessed 21 March 2013) a

[44]      For a discussion in Wired News see URL below, (last accessed 21 March 2013).http://wired-vig.wired.com/news/politics/0,1283,36612,00.html

[45]      As it happens, the use of resources such as Anonymiser meant that resourceful New Zealanders could still access the site.

[46]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001), p 46.

[47]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001), p 304.

[48]      111 F Supp 2d 294 (SDNY 2000); 273 F 3d 429 (2d Cir NY 2001), p 331.

[49]      For some of the exploits of the early hackers including members of 2600 see Bruce Sterling The Hacker Crackdown: Law and Disorder on the Electronic Frontier New York, Bantam Books, 1992.

[50]      2000 WL 1160678 SDNY 2000 (17 August 2000), pp 12–13.

[51]      2000 WL 1160678 SDNY 2000 (17 August 2000), pp 21–22.

[52]      A content-neutral restriction may be permitted if it serves a substantial government interest, unrelated to the suppression of free expression and the regulation is narrowly tailored which “in this context requires . . . that the means chosen do not ‘burden substantially more speech than is necessary to further the government’s legitimate interests.’” Universal City Studios v Reimerdes and Corley 111 F Supp 2d 294 (SDNY 2000), 273F 3d 429 (2d Cir NY 2001) at 49; Turner Broadcasting System, Inc v FCC (1994) 512 US 622, 662, Regulation of expression is content-neutral if the regulation is justified without reference to the content of the regulated speech Hill v Colorado, 530 US 703, 720, 147 L Ed 2d 597, 120 S Ct 2480 (2000).

[53]      Universal City Studios v Reimerdes and Corley 2000 WL 1160678 SDNY 2000 (17 August 2000) at 70