1. EU-Wide Licensing for Online Rights; Orphan Works

The Information Society Directive has now been in force for almost a decade. Although far from revolutionary, it established a more harmonised framework for copyright and related rights in the information society.Footnote ¹¹ It harmonised three important rights; the rights of reproduction, communication to the public/‘making available’, and distribution.Footnote ¹² It also provides an exhaustive list of twenty permissible exceptions, and a single mandatory exception (for certain transient or incidental copies made as an inevitable result of digital technology).Footnote ¹³ It has provided a useful baseline. However, the enthusiasm with which consumers have welcomed the products and services made possible by digital technology has meant that there is still a good deal more to do before there is ‘a comprehensive framework for copyright in the digital single market’.Footnote ¹⁴ It is a very significant economic area. According to the Commission, the cultural and creative sector (which comprises published content such as books, newspapers and magazines, musical works and sound recordings, films, video on demand and video games) generates a turnover of more than €650 billion annually and contributes to two point six percent of the EU's GDP, employing more than three percent of the EU work force.Footnote ¹⁵

The Commission has been frustrated for some time by the fragmentation of online markets in the EU.Footnote ¹⁶ Its argument is that there is potential to increase authors' and creators' returns if works could be licensed and disseminated within a digital single market. However, this model ignores the traditional boundaries of collecting societies, which are nationally organised, so there has been resistance to it. The Commission has also sought to argue that EU-wide online licensing is a way of disseminating different cultures and their repertoires across the EU, and has tried to persuade the creative community that it represents an opportunity and not a threat. There has been some anxiety that large collecting societies will capture the international commercially successful repertoire from national collecting societies, making it difficult for smaller societies to compete.

In 2005 the Commission adopted a Recommendation intended to improve the EU-wide licensing of copyright for online services, proposing the elimination of territorial restrictions and customer allocation provisions in existing licensing contracts, while leaving right holders the possibility of tendering their repertoire for EU-wide direct licensing.Footnote ¹⁷ It was made clear that if this soft-law approach had no effect, that then a binding legislative measure would be considered. Public consultation was undertaken, to assess the development of the online music sector, in the light of the Recommendation. This revealed that the Recommendation did seem to have had an impact, and that there is now a nascent market for EU-wide licensing of music for online services. A variety of other licensing initiatives have been launched or announced, and other negotiations are in progress.Footnote ¹⁸ A public hearing on the subject was held in 2010.Footnote ¹⁹ The Commission has now said that it will submit proposals to create a legal framework for the collective management of copyright to enable multi-territorial and pan-European licensing.Footnote ²⁰

As part of its wider thinking on these issues, in October 2009 the Commission published what it termed ‘a reflection paper on the challenge of creating a European Digital Single Market for creative content like books, music, films or video games.’Footnote ²¹ This again emphasised the Commission's view that ‘a truly Single Market without borders’ for online content could produce a many-fold increase in retail revenues. The Commission also takes the view that illegal downloads ‘on a large scale’ could jeopardize the development of an economically viable Single Market for digital content. It advocates much more encouragement for legal cross-border offers.

One obstacle to such initiatives are so-called ‘orphan works’—works still in copyright, but whose copyright holders cannot be located to negotiate permissions. In America, the ‘Google Books’ project has been extremely controversial, and is still a matter of legal dispute. But there is considerable political pressure to make cultural resources widely available in digital form. Within Europe, the Commission's ‘Europeana’ portal has now been launched.Footnote ²² It contains over fifteen million digitised items from museums, libraries, archives and audio-visual collections throughout Europe. The Commission, recognising the need for common rules on digitisation, has proposed a Directive on orphan works. This would establish rules for identifying orphan works, provide for a generally accessible record of such works, and establish permissible uses of them. It would also ensure that a right holder could assert copyright, and terminate a work's ‘orphan’ status.Footnote ²³ So-called ‘out-of-commerce books’—books no longer sold or published—are also problematic for such projects. The Commission is currently working with stakeholders to produce model forms of licensing agreements for this category of copyright works, also.

2. Copyright Levies

Copyright levies are imposed by many countries on recording hardware and/or blank recording media. They were originally justified on the basis that it is not practical to license individual acts of private copying. Copyright levies are now increasingly applied to digital equipment and media as a form of compensation for right holders whose works are subject to private copying—an approach permitted by the Information Society Directive.Footnote ²⁴ There are concerns that such levies are not always fairly targeted or distributed. The sums of money involved are very significant.Footnote ²⁵ Very strong and divergent views are expressed on this subject, and repeated consultations have failed to resolve these differences. In February 2008 the Commission announced that it would re-launch the consultation process, with a view to finding a pragmatic and workable solution.Footnote ²⁶

A recent ruling by the European Court of Justice, on the definition of ‘fair compensation’ under the Information Society Directive, is likely to influence these discussions. Padawan involved a reference for a preliminary ruling concerning the interpretation of the concept of ‘fair compensation’ paid to copyright holders in respect of the private copying exception.Footnote ²⁷ Predictably, the Court insisted that the concept of ‘fair compensation’ must be interpreted uniformly throughout the EU, notwithstanding the Member States' power to determine matters of detail (such as levels of compensation, arrangements for financing and collection). The Court also held that ‘fair compensation must be calculated on the basis of the criterion of the harm caused to authors of protected works by the introduction of the private copying exception’. It is significant that the Court saw private copying as necessarily harmful: ‘Copying by natural persons acting in a private capacity must be regarded as an act likely to cause harm to the author of the work concerned.’Footnote ²⁸ The person causing this harm to the holder of the reproduction right is the person who reproduces it for private use without prior authorisation. Therefore, the Court reasoned, it is for that person to make good the harm related to that copying by financing the compensation to be paid to the right-holder.

Identifying this individual in all cases is clearly impractical. Recital 31 in the preamble to the Directive envisages a ‘fair balance’ between the rights and interests of the right-holders and those of users. The Court held that it is consistent with ‘fair balance’ to provide that people who have digital reproduction equipment, devices or media, and who make them available to private users or provide them with copying services, should finance the fair compensation. These people are in a position to pass this burden on to the private users concerned. However, the Court also held that there must be a link between the application of such a levy, and the deemed use of digital reproduction equipment for private copying. So indiscriminate application of the private copying levy, in particular with respect to digital reproduction equipment not made available to private users and clearly reserved for uses other than private copying (such as backing up data), is incompatible with the Information Society Directive.

The ruling in Padawan seems likely to add further fuel to the debate on levies. It also raises new questions, as to whether in practice any system could be targeted in the way the Court envisages. See, further, the difficulties faced by Stichting de Thuiskopie, the Dutch body responsible for recovering the private copying levy imposed by Dutch law. A German company, Opus, sold blank recording media online, via Dutch-language websites targeting consumers in the Netherlands. The effect of the contractual arrangements was that Opus did not pay the levy either in Germany or in the Netherlands—and this was reflected in the lower prices charged to consumers. The Court again stated that it was the final user who was in principle responsible for paying fair compensation, but it was open to Member States to establish a private copying levy chargeable to the people who make reproduction devices and media available to that final user. Furthermore, any Member State which has introduced a system of private copying levies is responsible for ensuring that authors harmed by private use on their territory are actually compensated.Footnote ²⁹ Again, this seems difficult to ensure in practice, given the ease with which such products may be posted from other jurisdictions. However, the Commission fully recognises the difficulties, and is committed to efforts to resolve outstanding issues, with a view to legislation action at EU level ‘by 2012’.Footnote ³⁰

3. Copyright Term Extension—Sound Recordings

The term of protection for recorded music is currently fifty years. Almost a decade ago, there were calls to extend the term of this neighbouring right. At that time the Commission was not persuaded of the need for change. The proposal was revived in 2009, when the European Parliament voted in favour of a draft Directive. But the proposal was blocked in the Council, with a number of Member States acknowledging the evidential weight of empirical studies undermining the assertions made in support of the proposal. Economic evidence indicates clearly that extension of copyright term would have no net incentivising effect.Footnote ³¹ However, the political climate has changed yet again, and a Directive extending the term of protection for performers and sound recordings to seventy years was adopted on 12 September 2011.Footnote ³² The primary aim is to enable performers to earn money for a longer period of time, given that they often do not have other regular salaried income. The Directive contains a ‘use it or lose it’ clause, which will allow performers to reclaim assigned rights if a record producer does not market the sound recording during the extended period. In a further attempt to support performers, record companies are required to set up a fund into which they will have to pay twenty per cent of their revenues earned during the extended period. The money from this fund will be used to help session musicians.

1. Unitary Patent Protection—Further Delays

Unitary patent protection has become something of a holy grail for the EU. This quest is driven by a stark economic case. Currently, to obtain patent protection for all twenty-seven EU Member States for twenty years, a company would have to spend in the region of €200,000.Footnote ³³ A large part of these costs would be translation costs and costs resulting from essential dealings with national offices. This is many times the cost of registration with other leading patent offices, such as the US and Japan. But yet again it is necessary to report serious difficulties in establishing a unitary patent within the Community. The attractions of a unitary patent are significant; lower costs, legal uniformity and certainty, efficiency, and the elimination of forum-shopping. However, some Member States remain opposed to the scheme, for their own reasons. Decades of negotiations have failed to resolve differences over translations and jurisdictional arrangements. Patentees, litigants, judges, and the Commission itself have repeatedly expressed their frustration at this state of affairs. Given the economic significance of patents, it is not sensible to abandon efforts to resolve the situation, intractable though it has been.

In April 2007, the Commission produced yet more proposals, and there were considerable efforts to move things forward.Footnote ³⁴ These were again unsuccessful. In July 2010, the Council proposed a new approach to translation arrangements, based on the very successful tri-lingual arrangements at the European Patent Office (‘EPO’).Footnote ³⁵ The proposal was that an applicant would be able to file its patent application in any language. A ‘European Union Patent’ would then have been granted in one of the three EPO languages (English, French, German) and the applicant would then have to file translations of the claims in the other two EPO languages. Translation of the full specification would only be required if the patent was litigated, in an EU language of the choice of the alleged infringer. Unfortunately, a number of Member States considered these proposals unworkable, and Italy and Spain were doggedly opposed to the trilingual proposal, preferring either a multi-lingual or monolingual (English) regime. An extraordinary Competitiveness Council in November 2010 failed to break the deadlock.Footnote ³⁶

Since unanimity could not be achieved, a significant number of Member States—with the support of the Commission—decided to proceed on the basis of ‘enhanced cooperation’.Footnote ³⁷ This unusual tactic has the effect of side-lining any Member States blocking progress. This approach would allow participating Member States to establish a patent through a single application which would be valid in all participating Member States. By February 2011 all Member States except Italy and Spain had indicated that they wished to be involved in the enhanced cooperation procedure, and the European Parliament agreed to the exercise. In March 2011, the Competitiveness Council authorised the launch of enhanced cooperation, and the Commission has now tabled proposals for implementing measures. Two Regulations will be required; one on the creation of unitary patent protection and a second on the applicable translation arrangements.Footnote ³⁸

Another crucial element in EU patent strategy ought to be an EU-wide patent litigation system. Litigants currently have to pursue their cases in the relevant national courts—often in parallel. This is costly and inefficient, and is liable to result in different decisions in different jurisdictions. From the point of view of users, appropriate jurisdictional arrangements are essential. Following the very considerable efforts of the European Patent Lawyers Association, an agreement for the creation of a European and Community Patents Court was drafted. It envisaged a court of first instance, an appeals court and a joint registry. Signatories would have included EU states, the EU itself, and non-EU countries party to the European Patent Convention. In 2009 the Council, at the request of the Commission, submitted the draft Agreement to the Court of Justice for an opinion on its compatibility with EU law. The Court took the view that the draft Agreement was not compatible with EU law.Footnote ³⁹ A major concern was that the proposed court would be outside the institutional and judicial framework of the EU. The Court saw the draft Agreement as depriving the EU Member States of the ability to cooperate with the Court by referral of questions of EU Law, and was unhappy that the proposed Court would not itself have had to refer such questions to the Court of Justice.

This opinion has created serious difficulties for the Commission. As a practical matter, if unitary patent protection is to be agreed, users of the system will expect there to be one single forum with the power to enforce and revoke these patents throughout the territory of all of the relevant Member States. And if litigants are to have confidence in a new system, it will be important to ensure that the quality of legal judgments is high. Further proposals have been drafted, for the creation of a unified and specialised court to handle both the current ‘bundled’ European patents, and the new unitary patents (if these are agreed).Footnote ⁴⁰ There is pressure to secure political agreement to these proposals by 2012, although practitioners express concern that the draft still needs a good deal of work.

1. Registration—the Absolute Grounds for Refusal

The efforts made to register marks on the boundaries of registrability offer a testament to the success of the regime. Registration is clearly regarded as a prize worth fighting for. There have been a number of recent cases concerned with distinctiveness. Trade marks which are devoid of any distinctive character may not be registered.Footnote ⁴³ This is to ensure that marks can fulfil their function as a guarantee of origin. Although all marks are judged on the same basis, it may be particularly difficult to show that a shape is distinctive in the necessary sense. The relevant public may not regard a shape as communicating anything at all about trade origin. In Lindt, the famous Swiss chocolate company was attempting to register the shape of its foiled chocolate bunnies, mice, and reindeer. The General Court ruled that these are typical shapes in which chocolate and chocolate goods are presented, particularly at Easter and Christmas. As such, Lindt's shapes were devoid of any distinctive character.Footnote ⁴⁴ This decision is consistent with previous case law, which indicates that a shape mark must depart significantly from sectoral norms before it is regarded as distinctive.

There is a new fashion for ‘positional’ marks. In X Technology Swiss the applicant sought to register a shape on the toe of a sock in a particular shade of orange. Registration was refused, and the Board of Appeal upheld that conclusion in a carefully reasoned decision. Consumers would pay only a low degree of attention when purchasing these cheap, everyday goods. They would think that the mark was dictated by aesthetic and functional considerations. It is common to colour parts of a sock, and the colouring might indicate reinforcement. The relevant public was not in the habit of perceiving the colour of the toe of a sock as an indication of origin, and would be more likely to perceive the feature as decorative. The mark was thus devoid of distinctive character. The General Court dismissed a further appeal.Footnote ⁴⁵ Similarly, another position mark—two curved lines of stitching on a pocket—was refused registration on comparable grounds.Footnote ⁴⁶

In Glaverbel the applicant sought to register the texture of glass, for sheets of glass for various building purposes. The General Court held that the texture of glass was devoid of any distinctive character, because it would be considered by the relevant consumers as one of the possible functional aspects of a specific type of glass.Footnote ⁴⁷ Similarly, in Freixenet, the General Court upheld OHIM's decision to refuse the registration of the surfaces of two wine bottles as trade marks. Freixenet, an important producer of sparkling wines, sought protection for the surfaces of a black matt bottle and of a clear bottle which, when filled, took on a frosted matt golden appearance. The General Court agreed that the marks were not distinctive. The appearance of the bottle alone (whether its shape or its surface) did not indicate its origin. Consumers would need a label to identify the manufacturer. This was, in effect, an unsuccessful attempt to protect a product's packaging without using any text or graphics.Footnote ⁴⁸

Shape marks, like other marks, must satisfy the general requirements for registrability, including distinctiveness. In addition there are specific exclusions intended to prevent the registration of marks which will give their proprietors a monopoly on technical solutions or functional characteristics of a product.Footnote ⁴⁹ The leading authority is Philips: if the essential functional characteristics of the shape of a product are attributable solely to the technical result, it may not be registered, even if that technical result can be achieved by other shapes.Footnote ⁵⁰ This approach was applied to prevent the registration of the Lego brick for various goods, including construction toys. The Cancellation Division found that each of the elements of the shape of the Lego brick, and the Lego brick as a whole, were necessary to obtain a technical result. The various features (studs, secondary projections, hollow skirt, overall brick shape, etc) all performed particular technical functions (‘clutch power’, fixing versatility and arrangement, size for children, etc). On appeal, the Grand Board of Appeal agreed that the Lego brick's features were clearly adopted to perform utilitarian functions, and there was nothing arbitrary or ornamental present in it. Any ‘eye appeal’ merely stemmed from the aesthetics of a sound structural and functional form. Thus, in the words of Philips, ‘the essential functional characteristics of the shape’ were ‘attributable solely to the technical result’.

The General Court upheld this decision, and in doing so rejected Lego's argument that the essential characteristics of the shape at issue ought to be determined from the point of view of the consumer, taking into account consumer surveys. The General Court stressed that the aim of the provision is to assess the functionality of the shape at issue, and the perceptions of the target consumer are not relevant to this analysis—the essential characteristics of a shape must be determined objectively.Footnote ⁵¹ The Court of Justice dismissed Lego's further appeal. The Court emphasised that the correct application of the provision requires first that the essential characteristics of the shape must be properly identified—meaning, the most important elements of the sign. The competent authority then has to ascertain whether these essential characteristics all perform the technical function of the goods at issue. If the shape is one in which a non-functional element (decorative or imaginative, for example) plays an important role, then competitors will have ready access to alternative shapes with equivalent functionality, and there is no risk that the technical solution will be unavailable to them.

Another famous toy, Rubik's Cube, was also the subject of a shape trade mark application. Of interest here was the Board of Appeal's insistence that the assessment of its registrability had to be based on the representation of the mark as filed, and not on any alleged invisible features. Here, the representation consisted simply of three perspectives of a three-by-three-by-three cube, with each square facet separated from neighbouring facets by a black contour. Prima facie, the representations were not suggestive of any particular function, even when the goods—three-dimensional puzzles—were taken into account. It was impermissible to consider the well-known rotating capability of the Rubik's Cube, and thus illegitimately read the functionality back into the representations. As shown, the cubic grid structure did not give any indication whatsoever as to its function, or even as to whether it had any structure. It was therefore impossible to conclude that it might impart some technical advantage or effect in the domain of three-dimensional puzzles.Footnote ⁵²

There has been an important ruling from the Court of Justice on the concept of bad faith, which will be discussed in this section because Member States may provide that a national trade mark may be refused registration if the application is made in bad faith. However, it should be stressed that for the CTM this is a ground of invalidity and not a ground of refusal.Footnote ⁵³ The case again involved chocolate bunnies, which have been marketed in both Austria and Germany since at least 1930. The individual shapes of chocolate bunnies differed considerably when they were manufactured and wrapped by hand, but industrially manufactured bunnies have become increasingly similar. Lindt & Sprüngli has produced a chocolate bunny since the early 1950s, and in 2000 registered a three-dimensional mark protecting it. Franz Hauswirth has marketed chocolate bunnies since 1962, including one which, in the opinion of the referring court, would result in a likelihood of confusion with the Lindt bunny. Lindt had previously only taken action against manufacturers of identical bunnies, but after registration of its mark as a CTM, it relied on the terms of the Regulation to pursue manufacturers of products likely to cause confusion. Hauswirth counterclaimed that the mark was registered in bad faith, and the referring court sought guidance on the concept.

The Court of Justice held that the national court must take into consideration all factors relevant to the particular case, assessed at the time the application was filed. Some examples of relevant factors were given by the Court. It is relevant that the applicant knows or must know that a third party is using an identical or similar sign for an identical or similar product capable of being confused with the sign which the applicant is seeking to register. However, even if this is the case, this fact in itself does not amount to bad faith. The applicant's intention in registering the mark is also relevant, particularly if there is evidence that the applicant deliberately sought to use the registration to prevent third parties from marketing their products (perhaps, by registering a mark without intending to use it). If a third party has been using a sign for a considerable period of time (as here), and that sign did have some legal protection, then this could be relevant to the assessment of bad faith. For instance, the applicant could have the sole aim of competing unfairly with the third party. On the other hand, the Court stressed, an application in such circumstances could still have a legitimate objective; say, if the applicant's priority is to protect itself against new entrants to the market who are known to be copying its get up. Additionally, the nature of the mark applied for may be relevant. For example, if the sign concerned consists of the entire shape and presentation of a product, the fact that the applicant is acting in bad faith might more readily be established where competitors’ freedom to choose the shape of a product and its presentation is restricted by technical or commercial factors. In such circumstances the trade mark proprietor could prevent competitors not merely from using an identical or similar sign, but also from marketing comparable products. The Court's flexible approach leaves to the national court the challenging task of assessing the relevance of these and other factors, given the particular facts of the case.

2. Marks with a Reputation

Marks ‘with a reputation’ enjoy broader protection than others.Footnote ⁵⁴ A mark has a ‘reputation’ if a significant part of the relevant public has sufficient knowledge of it that, when confronted by a later trade mark, it may make an association between the two trade marks, even when used for non-similar products or services. In General Motors, a case concerning a national trade mark, the Court of Justice held that the mark's reputation need not cover the entire national territory, but it must exist in a substantial part of it.Footnote ⁵⁵ More recently, in PAGO, the same approach was taken in respect of a CTM.Footnote ⁵⁶ PAGO held a CTM for fruit juices. The essential element of this mark was the representation of a green glass bottle with a distinctive label and lid. In Austria PAGO sold a fruit juice called ‘Pago’ in exactly those bottles, so its mark was widely known there. Tirolmilch, a local Austrian competitor, sold a fruit and whey drink called ‘Lattella’ there. That drink, initially sold in cartons, was later packaged in glass bottles which in some respects resembled those in PAGO's trade mark. PAGO sued Tirolmilch for trade mark infringement, claiming that it was taking unfair advantage of the reputation of Pago's mark in Austria without due cause. The Austrian court wondered whether a CTM could be protected in the whole Community as a ‘trade mark with a reputation’ if it had a reputation only in one Member State. The General Court held that the mark must be known by a significant part of the relevant public, in a substantial part of the territory of the European Community. It added that Austria should be considered as constituting a substantial part of the territory of the Community. From one perspective, PAGO can be regarded as a welcome decision, because it ensures that a CTM is protected in a Member State to the same extent that a national mark would be. But it does not determine whether a single Member State will always be regarded as a substantial part of the Community. This might offer an opportunity for a canny litigant to establish a mark's reputation in one (smallish) country, and then use this to attempt to enforce the CTM in a different State—one in which it does not enjoy such a reputation.

To trigger protection, use of the later mark must take unfair advantage of the earlier mark, or be detrimental to its distinctive character or repute. The existence of a ‘link’ between the marks is an essential condition; the relevant public must make a connection between them. In the absence of such a link, the use of the later mark is not likely to take unfair advantage of, or be detrimental to, the distinctive character or the repute of the earlier mark. The existence of such a link must be appreciated globally, taking into account all factors relevant to the circumstances of the case. Those factors, listed in Intel, include: the degree of similarity between the conflicting marks; the nature of the goods or services for which the conflicting marks were registered, including the degree of closeness or dissimilarity between those goods or services, and the relevant section of the public; the strength of the earlier mark's reputation; the degree of the earlier mark's distinctive character, whether inherent or acquired through use; the existence of the likelihood of confusion on the part of the public.Footnote ⁵⁷ There will necessarily be a link between the conflicting marks if there is a likelihood of confusion. Distinctiveness and reputation are assessed by reference to the relevant public. Strongly distinctive marks will find it easier to show detriment. There is no need to demonstrate actual and present harm, but there must be a serious risk of unfair advantage or detriment. It is not necessary to prove intention to take unfair advantage, or to damage distinctive character or repute.

Under the Intel criteria, even a first use of the later mark may suffice to be detrimental to the distinctive character of the earlier mark. Somewhat controversially however, the Court of Justice has stated that ‘proof that the use of the later mark is or would be detrimental to the distinctive character of the earlier mark requires evidence of a change in the economic behaviour of the average consumer of the goods or services for which the earlier mark was registered consequent on the use of the later mark, or a serious likelihood that such a change will occur in the future.’ The earlier user need not show that the later user has obtained commercial benefit from the use of the earlier mark. But a strict application of this approach would seem to make it hard for trade mark owners to prove their case. The Court did not explain what evidence could be marshalled to prove that a change in economic behaviour has been caused by a reduction in the distinctiveness of their marks, rather than external factors such as, say, the general financial climate. The Board of Appeal has thus far taken a reasonably pragmatic view of this condition.Footnote ⁵⁸

Intel seemed quite tough on superbrands, in terms of what it demanded regarding proof of detriment to distinctive character. Shortly afterwards, in L'Oréal v Bellure, the Court of Justice considered the more narrow question of what constituted unfair advantage.Footnote ⁵⁹ This case seems to offer superbrands more comfort. At issue were ‘smell-alike’ perfumes, which were sufficiently different from their famous trade marked counterparts to avoid likelihood of confusion, but were marketed in a way that ‘winked at’ L'Oréal's famous perfume brands. L'Oréal was objecting to Bellure's use of comparison lists for each of their ranges of product, showing which Bellure product corresponded to which L'Oréal perfume. The evidence was that Bellure obtained a major promotional advantage from using such lists; neither customers nor ultimate consumers were deceived as a result of the use of the lists; neither the image nor the distinctiveness of the trade mark for the comparable fine fragrance was impaired by the use of the lists; sales of the corresponding fine fragrance were not affected by the use of the lists. The Court took the view that the referring court was asking whether the proprietor of a well-known mark can oppose use of an identical sign where that use is not capable of jeopardising the mark or one of its functions but none the less plays a significant role in the promotion of the goods or services of the third party. It held that taking unfair advantage of the distinctive character or the repute of a mark does not require that there be a likelihood of confusion or a likelihood of detriment to the distinctive character or the repute of the mark. It stated that unfair advantage occurs where a third party uses a similar sign, seeking ‘to ride on the coat-tails of the mark with a reputation in order to benefit from the power of attraction, the reputation and the prestige of that mark and to exploit, without paying any financial compensation, the marketing effort expended by the proprietor of the mark in order to create and maintain the mark's image’.

The L'Oréal decision comes very close to introducing a tort of unfair competition; though only for marks with a reputation. Jacob LJ in the United Kingdom Court of Appeal expressed strong dissatisfaction with the implications of this approach:

Jacob LJ also noted a risk to free speech, which he argued should be protected in the marketplace as elsewhere: traders should be able to make honest statements about lawful products, and consumers should be able to make an informed choice. There was also criticism of the Court of Justice's reference to ‘the functions’ of the marks in questions, ‘such as, in particular, their functions of communication, investment or advertising’.Footnote ⁶¹ As the Court of Justice has repeatedly stated, the ‘essential function’ of a trade mark is to indicate origin. The status of these other functions remains unclear, as does the level of legal protection they enjoy.

3. Infringement—Keywords

Consumers are now adept at using the internet to research and source purchases. The content of these searches, and the results returned, constitute commercial information of considerable value. Advertisers are willing to pay search engines for ‘keywords’, in order that their sponsored links will be displayed prominently, in addition to the ‘natural’ search results. Hotly disputed is whether sales of keywords by search engines should constitute infringement, if the keyword is a trade mark. Both legitimate competitors and counterfeiters might have an interest in buying a particular trade mark as a keyword. Different search engines will offer different features, designed to make their search product more attractive to advertisers. Google's product is called AdWords, and is the company's main source of revenue; Google's total advertising revenues were $28 billion in 2010. The Court of Justice has now handed down an important decision on the subject of keywords, Google France; three joined cases involving the AdWords product.Footnote ⁶² The referring national court had established that entering the claimants' trade marks into Google's search engine triggered the display of advertisements for sites offering counterfeit versions of the company's products. Google offered advertisers the possibility of selecting these trade marks as keywords both on their own, and in combination with expressions such as ‘imitation’, ‘replica’ and ‘copy’. These facts led to Google being held liable for trade mark infringement in the French courts. Google's appeal led to the referral of various questions for a preliminary ruling.

The Court of Justice considered the case where an advertiser is using a keyword identical to a trade mark, which has been selected through an internet referencing service (such as Google) without the trade mark proprietor's consent, and is using it for identical goods and/or services. In these circumstances, the trade mark proprietor is entitled to prohibit that advertising if the advertisement does not enable an average internet user to ascertain (or to ascertain only with difficulty) whether the goods or services referred to originate from the trade mark proprietor or a third party. There is criticism of this approach, because it seems to add a further requirement to the legislative provision, which simply requires identity of marks, and identity of goods/services. The Court uses the word ‘misleading’ to encapsulate the concept which triggers a proprietor's ability to object, and this arguably imports a requirement of confusion into a provision whose text does not require it. These facts also engage the Ecommerce Directive, article 14 of which provides an immunity for an Internet Service Provider (ISP) which is merely hosting someone else's information. There is an obligation to act expeditiously to remove unlawful data once knowledge is obtained. The Court applied this rule to the Adwords situation and concluded that it applies to an internet referencing service provider (such as Google) where it has not played an active role of such a kind as to give it knowledge of, or control over, the data stored.

Both sides have claimed the Court's ruling as a win. Google presents it as an affirmation that selling Adwords is legal, and that their service is (in principle) covered by the Ecommerce hosting safe haven. Trade mark proprietors emphasise the qualification that if the guarantee of origin is undermined, or if the ISP has actual knowledge of an advertiser's unlawful activities, then Google will be vulnerable. One clear effect of the Adwords ruling was that Google changed its keyword policy for all of Europe (aligning the United Kingdom and Ireland with the Continent)—making it on balance less favourable to trade mark owners. Unless they can show confusion, trade mark holders will have to live with seeing their marks used as Google Adwords, and will have to compete with counterfeiters and other rivals for their own trade marks. There are a real policy issues here. Brand owners—particularly luxury brand owners—want to control their brand's image, reputation and status. There is unquestionably a huge volume of counterfeiting. However, some commentators consider that trade mark proprietors seek an absolute right of control which is not justified, and not in the interests of consumers. In Google France the Court has attempted to steer a course between the interests of ISPs and trade mark owners. Another recent ruling on the subject, Interflora,Footnote ⁶³ addresses the effect of keyword sales on marks with a reputation. It will be interesting to see what the effects of these rulings are in practice.

1. Designs

The design of a product is often of great importance to consumers, who may well choose their purchase on this basis. This leads to demands for design protection. Legislatures have responded by using many different models, sometimes simultaneously. The ‘patent’ approach involves registration, perhaps pre-examination, and provides a degree of legal certainty. The ‘copyright’ approach does not require formalities or fees, but its accessibility comes at the expense of certainty regarding protection. The success of the Community's Registered Community Design shows that where the process is made cheap and easy, registration will be a popular route; about four hundred and sixty thousand designs have so far been registered, and this continues at the rate of about seventy-five thousand per year. There is no substantive examination, and many designs can be registered within a week. Also on offer is the Unregistered Community Design. In both cases, eligible designs must be new and must have an individual character. A Registered Community Design is protected against both deliberate copying and the independent development of a similar design. It lasts for up to twenty-five years. The Unregistered Community Design offers short-term protection against deliberate copying, for three years from the date of disclosure of designs to the public within the EU.

OHIM's case law continues to develop, and there have been some significant decisions from the General Court. In PepsiCo, PepsiCo applied to register a design for a ‘tazo’ or ‘rapper’—a pog (or token), often included in crisp packets as a collectible promotional item. After an objection from a Spanish company, Pepsi's design was initially declared invalid. The OHIM Board of Appeal reversed this decision, because the designs produced a ‘different overall impression’ on the ‘informed user’, and so did not infringe under the terms of the legislation. The matter was appealed to the General Court, which noted that the question of whether or not a different overall impression is produced requires consideration of the degree of freedom of the designer in developing the design. If this freedom is restricted, minor differences between the designs may suffice to produce a different overall impression. Here, ‘rappers’ are necessarily curved, so that when they are pressed in the centre a sound is made. This severely restricts the freedom of the designer. The General Court considered that the informed users would focus on features that were arbitrary, or different from the norm. Here, the differences between the two designs were too minor to produce a ‘different overall impression’ on the ‘informed user’. Thus, the PepsiCo design application was indeed invalid.Footnote ⁶⁴

2. Geographical indications

The Community has a well-established system for the protection of geographical indications and designations of origin for agricultural products and foodstuffs. However, there is currently no such system for the protection of non-agricultural products; such as Carrara marble or Solingen knives. The Commission is concerned that this leads to an un-level playing field in the Single Market, and has recently announced plans to review the existing national frameworks with a view to possible legislation at EU level.Footnote ⁶⁵