Technology, Media &

Telecommunications. Data Protection & Freedom of Information Contents

Asia Data Protection & Freedom of Information 1 IT & E-Commerce 12 Google Becomes an Advocate of Privacy Media & Broadcasting 23 During a UNESCO conference in September on the Internet and ethics,

Peter Fleischer, head of privacy at Google, promoted the adoption of a unified internationally-applicable set of privacy rules in order to protect personal data online.

A key reason for this is that the OECD’s guidelines (which were adopted in 1980) and the EU Data Protection Directive (which was adopted in 1995) pre-dated the widespread adoption of the Internet and are now failing to cope with modern privacy challenges. In their place, he focused on APEC’s Privacy Framework suggesting that all countries adopt the Privacy Framework agreed between Asia Pacific nations and endorsed by APEC Ministers in November 2004.

International Complexities

Google’s new interest for privacy may be surprising at first given its recent run-in with the Article 29 Working Group, who urged it to shorten the retention period of personal data in its server logs (such as IP number, web browser being used, date and time of connections), see Article 29 Working Party Puts Google Under The Spotlight.

However, the lack of international standards for data protection, particularly in respect of international data transfers, is a serious concern for Google as the company’s data centres are located all over the world and are subject to massive data flows. Accordingly, the adoption of global standards could facilitate the search engine’s compliance with privacy laws across borders.

Is the APEC Privacy Framework Fit For Purpose?

The choice of the APEC Privacy Framework as an international standard for data protection will probably be challenged by some. The APEC Privacy Framework has a set of nine privacy principles:

• preventing harm

• uses of personal information

Issue 45 ⏐ 1 November 2007 1

• notice

• collection limitations

• choice

• integrity of personal information

• access and corrections

• security safeguards, and

• accountability.

However, this omits a number of key data protection principles from the EU Data Protection Directive, such as legitimacy, specific restrictions on sensitive data, registration of data controllers and the appointment of a national regulator.

Moreover, experts already say that the APEC standards are too lenient and do not limit data collection to legitimate and specific purpose, nor subject data flows to an adequate level of protection. Both of these are key issues when considering Google’s data protection practices, in particular with regard to its recent acquisition of the web advertising network DoubleClick. In addition, if Spanish and French data protection regulators seem to have approved the idea of a global set of rules, Alex Türk, president of the French CNIL, nevertheless pointed out that the harmonization process could be quite long because of the different approaches to privacy among Europe, the United States and Asia Pacific nations and the lack of national regulation in some other countries.

The fact that the APEC Privacy Framework focuses on whether a person is harmed by a company’s privacy practices, rather than on violation of privacy as a right, will probably be a major sticking point in any move to a global consensus.

By Eléonore Feld, Paris

Belgium

Draft Regulation for Blacklists

Blacklists are widely used in all types of organisations and sector of activity. Draft regulations have been prepared that will set out strict criteria that must be met in order for these lists to be legal.

European Position

A generic definition of blacklists is set out in the Article 29 Working Party’s Working Document on Blacklists (WP 65) adopted on 3 October 2002:

“a blacklist could be said to consist of the collection and dissemination of specific information relating to a specific group of persons, which is compiled to specific criteria according to the kind of blacklist in question,

Issue 45 ⏐ 1 November 2007 2

which generally implies adverse and prejudicial effects for the individuals included thereon and which may discriminate against a group of people by barring them access to a specific service or harming their reputation.”

This working document recognised that inclusion in this type of list can have serious consequences for individuals and that there are clear discrepancies in the way these files are registered and used in Member States.

Belgium Proposes a Draft Royal Decree

In Belgium, a draft Royal Decree regulating blacklists is currently under review. Its purpose is to make the processing of such lists subject to much stricter controls.

The main new conditions imposed by the draft Royal Decree are:

• Enhanced notification - A data controller must file a notification with the Belgian Privacy Commission before starting any wholly or partly automated data processing operation. However, in the case of negative lists, such notification is enhanced since the data controller is required to include specific additional information regarding the blacklist in the declaration form.

• Authorisation - According to the draft Royal Decree, prior authorisation from the Belgian Privacy Commission would be required for all external negative lists, i.e. lists shared between several companies.

• Data protection officer - Finally, the data controller must appoint a data protection officer to supervise the processing of any external blacklists.

Belgian Privacy Commission’s Opinion

The Belgian Privacy Commission has already commented on the draft Royal Decree in an opinion dated 4 July 2007. Its main concerns are the following:

• There is no legal basis to grant the Privacy Commission the power to participate in the blacklists authorisation process. Moreover, the financial impact for the Commission of putting such a procedure in place has not been taken into account.

• The designation of a data protection officer should not be limited to external blacklists but should be extended to all types of blacklists.

Future Developments

Even though the draft Royal Decree may well be amended in light of the Privacy Commission’s concerns, organisations must keep in mind that it is very likely that the use of blacklists will be strictly regulated in Belgium in the near future.

By Sylvie Rapoport, Brussels

Technology, Media & Telecommunications

United Kingdom

ICO Guidance on Personal Data: Clarification or Further Confusion?

In August 2007, the UK Information Commissioner issued fresh guidance on the meaning of personal data (the “ICO’s Guidance”). This appears to be an attempt to reconcile the unorthodox position under English law following the Durant v Financial Services Authority [2003] EWCA Civ 1746 case with the Article 29 Working Party’s recent opinion on the mainstream EU approach to the concept of personal data (the “Working Party’s Opinion”).

This article considers how successful this exercise has been and the impact the ongoing regulatory tension will have in practice.

The issue is timely given that there are ongoing discussions between the European Commission and the UK Government in relation to various alleged infringements by the UK relating to the implementation of the Data Protection Directive. The Commission appears to be concerned both with the core UK legislative framework, i.e. the Data Protection Act 1998, and also the subsequent interpretation of that Act by the UK courts. One of the specific issues, in relation to which the UK appears to be most out of line with other EU member states, is indeed the scope of the concept of “personal data”.

English Law - Durant

The definition of personal data was considered in some detail by the Court of Appeal in Durant. The Court of Appeal’s conclusions are ably summarised in a later decision of the High Court, Smith v Lloyds TSB Bank Plc [2005] EWHC 246:

“(a) .. not all information retrieved from a search against an individual's name or unique identifier is personal data within the 1998 Act, (b) .. mere mention of an individual in a document held by a data controller does not mean that the document contains personal data in relation to that individual, (c) .. whether information is capable of constituting personal data depends on where it falls in a continuum of relevance or proximity to the data subject, (d) .. in answering that question it is relevant to consider whether the information is biographical in a significant sense; and whether it has the putative data subject as its focus and, finally, (e) .. personal data is information that affects the privacy of the putative data subject, whether in his personal, business or professional capacity.”

This narrow approach has been subject to some criticism and it is widely believed that it is one of the main drivers behind the European Commission’s broader investigation into the UK’s implementation of the Data Protection Directive.

However, the approach is arguably not without its merits, at least from a pragmatic perspective. In the digital age vast amounts of ‘unstructured’ electronic information is generated about individuals and approaching the definition of personal data in the manner adopted by the Court of Appeal in Durant increases the likelihood that the Data Protection Act will be focussed on cases where there is a real risk to an individual’s privacy.

Issue 45 ⏐ 1 November 2007 4

The Working Party’s Opinion

In contrast, the Working Party’s Opinion on the concept of personal data (WP 136) takes a very broad approach to this definition so as to prevent any “shadow zones” within its scope. The reasoning for this is that any difficulties raised by a broad definition can be dealt with by applying the other rules within the Data Protection Directive flexibly and proportionately.

With this in mind, it recommends a four stage approach to interpretation:

• Step 1: Is it information? This is interpreted widely and includes both objective and subjective information.

• Step 2: Does it relate to a person? This is one of the more controversial aspects of the Working Party’s Opinion. The restrictive approach taken by the English Court of Appeal is comprehensively rejected and replaced with the proposal that information can relate to a person through its (a) content, (b) purpose or (c) result.

• Step 3: Is that person identified or identifiable? This again is interpreted widely so that the person may be identified directly (e.g. through a name) or indirectly (e.g. through a combination of factors recorded about them). Similarly, the concept of being ‘identifiable’ is considered broadly.

• Step 4: Is the person a living natural person? While less contentious, the Working Party’s Opinion does consider some unusual examples such as unborn children and frozen embryos.

The Working Party’s Opinion is also entirely silent on the issue of “filing systems”, a topic of significant debate in the UK, at least following the decision in Durant.

The ICO’s Guidance

It appears that the ICO’s Guidance is an attempt by the Information Commissioner to reconcile the radically different approaches taken by the Court of Appeal in Durant and by the Working Party in its recent opinion. However, in both form and substance it is much closer to the Working Party’s Opinion in taking an expansive approach to the definition of personal data. As the preface to the ICO’s Guidance states: “We recognised the need to produce guidance with a greater emphasis on what is covered than what is not”.

The ICO’s Guidance is structured around a ‘flowchart’, set out in table 1. The first question considers if the individual is identifiable (broadly comparable to the “Third Element” of the Working Party’s Opinion), while the remaining seven questions, making up the bulk of the ICO’s Guidance, concentrate on whether the information “relates to” the individual (broadly comparable to the Second Element of the Working Party’s Opinion). The ICO’s Guidance does not consider what constitutes information or who are natural living persons in

Technology, Media & Telecommunications

any depth (see the First and Fourth Elements of the Working Party’s Opinion respectively).

It seems sensible to focus on whether information “relates to” an individual as it is the most difficult and controversial part of the analysis. The ICO’s Guidance transposes the ‘content’, ‘purpose’ and ‘result’ concepts in the Second Element of the Working Party’s Opinion by considering:

• if the information is ‘obviously about’ the individual (question 3) - i.e. is the ‘content’ of the information about the individual;

• if the information is used to inform or influence actions or decisions affecting the individual (question 4) - i.e. the information is collected with the ‘purpose’ of affecting an individual; and

• if the information impacts or has the potential to impact an individual (question 8) - i.e. does the information ‘result’ in an impact on the individual.

These issues are also interspersed with concepts from Durant, so that it is also necessary to consider if the information has biographical significance (question 6) or if the individual is the focus of that information (question 7). These questions do not feature in the Working Party’s Opinion. However, neither question allows one to conclude that the information is not personal data, so neither in any way narrows the scope of this definition.

Again, the ICO’s Guidance does not look at “relevant filing systems”. This will be the subject of further guidance in the near future.

Binding Precedent

The ICO’s Guidance does not appear to be particularly loyal to Durant. Some of the more obvious problems include:

• the lack of any link to an individual’s privacy. In Durant, Auld LJ’s analysis of the term personal data ends by stating “In short, it is information that affects his privacy, whether in his personal or family life, business or professional capacity”. This is not reflected in the ICO’s Guidance;

• the treatment of the ‘biographic’ and ‘focus’ questions. Both of these are identified in Durant as notions that may be of assistance when determining whether information is personal data. While these questions are part of the ICO’s Guidance, neither notion allows one to conclude that the information is not personal data. This does not seem to be consistent with Durant;

• the examples in the ICO’s Guidance. One example considers personal data in complaint files. If an individual complains that a request under the Freedom of Information Act 2000 is not being dealt with properly then the file about that complaint is likely to be personal data. This again seems to contradict Durant in which Auld LJ stated:

Issue 45 ⏐ 1 November 2007 6

“Just because the … investigation of the matter emanated from a complaint by [Mr Durant] does not, it seems to me, render information obtained or generated by that investigation, without more, his personal data”.

The status of the ICO’s Guidance is therefore somewhat difficult. In England, the formal interpretation of legislation is the exclusive function of the courts and in Durant the term “personal data” was authoritatively construed by a unanimous decision of the Court of Appeal. Therefore the Information Commissioner would make an error in law if he acted inconsistently with this judgment, entitling the court to quash any such decision or order.

The ICO’s Guidance will also have little impact on any future decision of the English courts as, being part of a common law system, they are bound by stare decisis to follow the precedent set in Durant. This principle binds both the High Court and, by and large, the Court of Appeal itself, to follow the decision in Durant.

Clarification or Further Confusion?

The ICO’s Guidance follows not only the substance of the Working Party’s Opinion but also its form and, even though it is categorised as “Technical Guidance”, it is unlikely to be particularly accessible as a practical guide even for a professional reader, such as a privacy officer. At 21 pages long it is twice the size of the Information Commissioner’s initial guidance on the definition of personal data (which also managed to cover the definition of “relevant filing systems”) and is much less “user-friendly” than other recent guidance from the Information Commissioner’s office.

However, the more important issue is how practitioners should treat the ICO’s Guidance in practice, especially given that an apparent gulf remains between the approach taken by the Court of Appeal in Durant, on the one hand, and by the ICO and the Article 29 Working Party, on the other?

While some aspects of the ICO’s Guidance may remain largely of academic interest (such as the example of the estate agent who takes a picture of a property at the same time as a nearby bank raid) others have quite significant practical implications, especially when deciding if “business information” should be disclosed in response to a subject access request.

Given the legal questions raised by the ICO’s Guidance, and the ongoing dispute between the UK Government and the European Commission over the concept at the heart of the document, data controllers may be tempted to treat this latest guidance with caution for the time being.

This article was originally featured in issue 7(9) of World Data Protection Report

The new guidance is available here.

By Christopher Millard and Peter Church, London

Technology, Media & Telecommunications

Table 1 - ICO Guidance

The information is not personal data

yes

yes

no

Q1. Can a living individual be identified from the data, or, from the data and other information in the possession of, or likely to come into the possession of, the data controller?

Q2. Does the data ‘relate to’ the identifiable liv g individual, whether in personal or fa y life, business or profession?

inmil

yes

no

no

unsure

Q3. Is the data ‘obviously about’ a particular individual?

Q4. Is the data ‘linked to’ an individual so that it provides particular information about that individual?

The information is personal data

Q5. Is the data used, or is it to be used, to inform or influence actions or decisions affecting an identifiable individual?

Q6. Does the data have any biographical significance in relation to the individual?

Q7. Does the data focus or concentrate on the individual as its central theme rather than on some other person, or some object, transaction or event?

Q8. Does the data impact or have the potential to impact on an individual, whether in a personal, family, business or professional capacity?

no

yes

yes

no

no or unsure

yes*

yes*

no or unsure

yes no*

Issue 45 ⏐ 1 November 2007 8

* Indicates that this is a likely, rather than a definite, conclusion

Freedom of Information: Widening the Net?

The Government made a series of announcements on 25 October 2007 to bolster the Freedom of Information Act 2000 (“FOIA”), enhance openness and increase public access to information.

These include a proposal to make additional organisations, from both the public and private sectors, subject to FOIA. It has also abandoned amendments to the charging regime that would have severely curtailed the impact of FOIA.

Who is Caught by FOIA?

Under FOIA any person can request information from “public authorities”. A public authority is one that is:

• listed in Schedule 1. This includes a vast range of public sector and quasi-public sector bodies from the Financial Services Authority to Sir John Soane's Museum;

• wholly owned by the Crown or a public authority listed in Schedule 1; or

• designated as a “public authority” by the Secretary of State because it (a) appears to the Secretary of State to “exercise functions of a public nature”, or (b) is “providing under a contract made with a public authority any service whose provision is a function of that authority” (section 5 of FOIA).

Outline of the Consultation

The Government has now issued its long-awaited consultation on the third limb of this test - whether any additional organisations should be designated as a public authority.

Such a designation could be a heavy burden for an organisation, both because it may be obliged to disclose information about its operations and also in terms of the additional administrative overheads in responding to requests for information.

Mindful that a designation may be very unpopular, and could be challenged in the courts, the consultation looks carefully at which organisations are likely to meet the relevant criteria in section 5. The consultation lists the following factors as relevant when considering if an organisation “exercises functions of a public nature”:

• the extent to which that function is publicly funded;

• whether the function is underpinned by statute;

• if the organisation has extensive or monopolistic powers;

• whether it is mandatory for parties to submit to that organisation’s jurisdiction; and

Technology, Media & Telecommunications

• whether the organisation acts for the public benefit or participates in a significant way, in the affairs of the nation.

The consultation also considers when an organisation is “providing under a contract made with a public authority any service whose provision is a function of that authority”. In looking at this issue it is necessary to look carefully at the service in question to see if it really is a function of a public authority (such as running children’s homes) or just a service that is just used internally by that public authority (such as the supply of stationery or IT equipment).

The power to make a designation is discretionary, so the consultation goes on to consider whether it will be appropriate to use this power. Key factors include the extent to which that organisation is funded by the public and the potential burden imposed by a designation. In relation to service providers, the consultation also recognises that there will already be a substantial degree of accountability arising out of the contract with the public authority.

Who might be Caught?

The consultation does not directly recommend that any organisation or class of organisation should be designated a public authority under section 5 of FOIA. However, based on the approach in the consultation, a number of organisations may well be subject to further scrutiny:

• organisations performing self-regulatory functions, such as the Law Society, Institute of Chartered Accountants or Bar Council. Each of these bodies has extensive powers, exercised in the public good and it is mandatory to submit to their jurisdiction;

• sports regulators such as the Football Association, the Rugby Football Union or the Lawn Tennis Association. Again, these organisations clearly have a public role in managing their respective sports, and it is mandatory to submit to their jurisdiction, giving them extensive quasi-monopolistic powers. Many also receive significant public or quasi-public funding. However, it is worth noting that a previous attempt to judicially review the Football Association failed indicating that its activities do not entirely fall within the public sphere (R v Football Association Limited ex parte Football League Limited [1993] 2 All ER 833.);

• major transportation operators such as Network Rail, Metronet or Tubelines. Network Rail will be particularly alert to this issue given the recent decision of the Information Tribunal that it does not perform “functions of a public administration” and therefore is not subject to the Environment Information Regulations 2004 (see Network Rail v Information Commission EA/2006/0061 & 0062);

Issue 45 ⏐ 1 November 2007 10

• PPP and PFI vehicles. In many cases these service providers have contracts with the Government for the performance of core Government functions such as the provision of hospitals, schools or roads. Designation of these entities would also make the application of FOIA more consistent, as access to information would depend on the service in question rather than the entity providing that service.

However, it is interesting that there is no reference to the utility companies exercising “functions of a public nature” and therefore being designated under section 5. This may well reflect the fact that, while these organisations were traditionally state run, they are now fully privatised and, in most cases, operating in a competitive market place. In these cases, there may be a stronger argument for the designation of historical records covering periods that such entities were in public ownership, publicly funded or operated as monopoly suppliers.

Retreat on Fee Regulations

The Government has also abandoned proposed amendments to the FOIA charging regime. Under these proposed amendments, a public authority could have included time spent considering the application of an exemption and associated public interest test when deciding if the request is too expensive to respond to (i.e. if the cost of dealing with the request would exceed the “appropriate limit”).

This could have led to more sensitive requests, such as the request for the Attorney General’s advice on the legality of the war in Iraq, being rejected solely on the basis that it would take too long to determine if the relevant information is exempt.

The Government received 324 responses to its consultation on the charging amendments of which 236 were against the proposals. Particularly surprising was that, of the 118 public authorities that responded, nearly half were against the amendments to the charging regime or undecided.

A Return to Openness

The freedom of information regime undoubtedly places an additional burden on public authorities and is sometimes misused. However, in the two and a half years since it was introduced it has helped to open up the workings of government. In Gordon Brown’s words:

“Freedom of Information can be inconvenient, at times frustrating and indeed embarrassing for governments. But Freedom of Information is the right course because government belongs to the people, not the politicians.”

The consultation is available here.

Responses must be provided by 1 February 2008.

By Richard Cumbley and Peter Church, London

Technology, Media & Telecommunications

IT & E-Commerce

European Union

Court Rules In Microsoft Case

In its 2004 decision, the Commission concluded that Microsoft had committed two types of abuse which where both related to the dominant position which it held through Microsoft Windows in the PC operating systems market: first, the Commission held that Microsoft had failed to supply its competitors with information necessary for them to allow their workgroup server operating systems to interoperate with Microsoft’s workgroup server and PC operating system; secondly, Microsoft had engaged in an illegal tie by bundling its Windows Media Player into its Windows operating system. The Commission imposed a record fine on Microsoft and imposed two remedies upon the company.

The Interoperability Case

In September 1998, Sun Microsystems called upon Microsoft to disclose all the information that it required to enable its servers to interoperate natively with Microsoft systems. Three months later, despite Microsoft’s offer of discussions about interoperability, Sun filed a formal complaint with the Commission alleging that Microsoft was guilty of an abusive refusal to supply the requested information. At the end of an investigation lasting over five years, the Commission concluded that Sun’s complaint was substantially justified and ordered Microsoft to prepare and license to third parties complete specifications for the protocols relating to interoperation between Microsoft Windows PCs and workgroup server operating systems (whether that interoperation was direct or indirect through Microsoft Windows server systems).

Although there was a dispute as to whether Microsoft’s refusal to disclose the information requested by Sun was actionable, there was no doubt that Microsoft would not meet the standard of disclosure required by the Commission. The obligation to disclose was based upon a factual assessment in which the most important elements were the findings that: (i) there were two relevant markets, an upstream market for PC operating systems and a downstream market for workgroup server operating systems (comprising systems that provided file and print and group and user administration functionality for servers costing less than $25,000); (ii) not only was Microsoft admittedly dominant on the upstream market (with a share in excess of 90%) but it had also become dominant on the downstream market (with a share that had grown to a level exceeding 60%);

Issue 45 ⏐ 1 November 2007 12

(iii) Microsoft had grown in the downstream market at the expense of rival suppliers, despite the superior quality of their products, because consumers attached substantial importance to a high level of interoperability; (iv) without intervention, other suppliers would be unable to achieve the level of interoperability required by customers and would be condemned to a condition of progressive marginalisation. In these circumstances, the Commission concluded that intervention was justified either on its preferred basis (that, on an assessment of all the circumstances, the benefits of ordering Microsoft to supply outweighed the detriments) or on the basis of the four-part IMS test (namely that (a) disclosure is indispensable to operation in a downstream market; (b) non-disclosure will result in the elimination of all competition in the downstream market; (c) non-disclosure will frustrate the development of a new product and (d) there is no objective justification for the non-disclosure).

At the heart of a wide-ranging defence that challenged all these assertions, Microsoft claimed that: (i) the Commission’s assessment was at odds with a market reality characterised by widespread interoperation amongst heterogeneous software systems; (ii) the mandated disclosure was not indispensable because there were alternative means available to rivals to achieve workable interoperability; (iii) the Commission had failed to give due weight to the successful entry of others such as, most notably, Linux and (iv) Microsoft was justified in refusing to disclose or license information that represented the fruit of substantial investment in which it had valuable intellectual property rights.

Although the Court did not adopt the Commission’s preferred legal standard, it confirmed the Commission’s decision on the basis of the four-part IMS test. In doing so, it stressed that: (i) the heterogeneity to which Microsoft referred was irrelevant because it lay outside the relevant market correctly identified by the Commission; (ii) it was essential that, within that market, a Windows client and server system should communicate with a third-party server as if the latter were a Windows system; (iii) that level of communication could not be achieved without the mandated disclosure by Microsoft; (iv) it was insufficient to show that niche suppliers might survive in competition with each other if they could not challenge Microsoft; (v) the mandated disclosure would not threaten Microsoft’s IPRs because it was limited to the specifications and did not extend to Microsoft’s implementation of those specifications; (vi) new products would be created because third parties would have to do that to meet customer demand and to avoid infringement of Microsoft’s IPRs; and (vii) Microsoft could not rely upon its IPRs to justify non-disclosure where the first three limbs of the IMS test were satisfied or upon unparticularised claims of threats to incentives to innovate, especially where non-disclosure was inconsistent with normal market behaviour and Microsoft’s own behaviour at a time when its share of the workgroup server market was small.

Technology, Media & Telecommunications

The Windows Media Player Case

Contrary to the Interoperability Case, the Commission’s story of harm and the underlying facts in the Windows Media Player case are relatively straightforward. Since the early 1990s, Microsoft had bundled a media player with Windows to which streaming functionality was added in 1999. Given that Windows was on more than 90% of all PCs, this guaranteed Microsoft’s Windows Media Player ubiquity in the market. According to the Commission, other forms of distribution, such as downloading, were less efficient and did not allow rival media players to achieve the same reach. The Commission took the view that this competitive advantage would ultimately lead to a tipping of the market and the elimination of competition media player providers: due to WMP’s presence on almost every PC, content providers would primarily encode their content in a Microsoft-compatible format and application vendors would write applications primarily for WMP. This, in turn, would make WMP more attractive to customers and trigger a virtuous circle which would ultimately lead to a Microsoft monopoly in media players with streaming functionality. In its support, the Commission pointed to declining market shares of RealPlayer, the dominant media player at the end of the 1990s, and rising market shares for WMP.

The Commission decided that Microsoft’s behaviour amounted to an illegal tie. Microsoft had bundled two separate products, its media player and Windows, its dominant PC operating system, and deprived customers (OEMs and end-users) of the choice to obtain Windows without media player, thereby restricting competition. Microsoft was accordingly ordered to supply a version of its PC operating system from which media functionality had been removed.

Microsoft, by contrast, argued that its behaviour did not restrict competition. Media players were offered for free by the various software vendors and there were several easy routes to market, for example, downloading. The open access to the market was evidenced by RealPlayer’s comparable presence on PCs. Furthermore, there was no evidence of “tipping” of the market: many content providers would provide their content in different formats and many software vendors would write their applications for different media players, which would allow the continued existence of different media players. In particular, the bundling of Windows and WMP should not be regarded as an illegal tie, as Windows and WMP should not be regarded as separate products for the analysis of tying and customers were in no way forced to use WMP, which furthermore was made available for free.

The Court confirmed the Commission’s decision that Microsoft’s bundling of Windows and WMP was an illegal tie. According to the Court, PC operating systems and media players were to be regarded as separate products for

Issue 45 ⏐ 1 November 2007 14

the tying analysis because there was demand for standalone media players and media players had different characteristics from operating systems. As a result of the bundling, customers were not given the choice to acquire Windows without simultaneously acquiring WMP and the fact that Microsoft did not charge for WMP and that customers were not forced to use WMP did not alter the Court’s conclusion. The Court concluded that the bundling with Windows gave Microsoft an unfair advantage which prevented competition between RealPlayer and WMP to be decided on the basis of the intrinsic merits of the two products. Finally, the Court rejected Microsoft’s efficiency justification on the basis that those efficiencies could either be obtained without the illegal tie (by means of OEMs’ decisions to integrate media players into their PCs) or could not be taken into account because they are the direct consequence of the impugned conduct.

Implications

The Microsoft judgment had been touted as the judgment of the decade rather more because it was perceived to be a test of the Commission’s enforcement credentials than because of its substantive importance for the future development of Community antitrust policy. Commissioner Kroes’ reaction to the judgment betrays the tension that even the Commission feels about the case, simultaneously describing the Commission’s decision as a landmark which sets an important precedent for dominant firms yet is in many ways exceptional because it is directed to an instance of super-dominance which is rarely observed.

Although the latter observation may seem reassuring, the Court’s judgment does have important implications both for dominant firms (and indeed for complainants) as well as for the future direction of EC competition policy.

Implications for dominant firms

At a time when the Commission has started a debate about the modernisation of Article 82 and has incorporated several modernisation elements into the Microsoft decision, the Court has turned back the clock and has reaffirmed the old analytical framework of the 1970s: the Microsoft judgment follows the long tradition of cases from Hoffman La-Roche to BA/Virgin and Michelin II which favours a formalistic approach in preference to a specific effects-based analysis and exercises great restraint in the scope of judicial review. If anything, the Microsoft judgment has further lowered the thresholds for a finding of abuse.

Form-based analysis: in its Microsoft judgment, the Court reiterated its statements in Michelin II and BA/Virgin (restated by the Court of Justice in the latter case) that conduct may be abusive if it is capable of restricting competition. That implies two questions: what is unrestricted competition and what level of restrictive capability is required to merit prohibition?

Technology, Media & Telecommunications

As to the first question, in both parts of the case the Court emphasised the importance that it attaches to competition based on the intrinsic merits of the rival products undistorted by any advantage that a dominant firm may derive from its operations in another market. If the advantage is large enough (as the Court clearly deemed Microsoft’s dominance in the PC operating system market to be), it may have to be shared with rivals. So, in the WMP case, the Court made it clear that the Commission’s detailed analysis into the effects of Microsoft’s bundling of Windows and WMP, while correct, was not required for a finding of abuse, nor was there any need to demonstrate consumer harm.

As to the second question, the Court stressed (reiterating the position stated in Tetra Pak II) that the Commission is not required to wait until competitors have exited the market before it intervenes. It is sufficient that such exit may be anticipated. The level of certainty that is required was effectively left open.

Limited judicial review: in merger cases, the Court has repeatedly intervened even in the context of complex appraisals and has assessed whether the evidence put forward by the Commission was factually accurate, reliable and consistent and whether it was capable of substantiating the conclusions drawn from it. In cases of abuse of dominance, by contrast, the Court has regularly emphasised that the complex appraisals by the Commission are subject only to a limited review by the Court. The Microsoft judgment confirms this bifurcated approach of the Court in competition cases. While paying lip service to the standard and burden of proof developed in merger cases, the Court repeatedly highlights the Commission’s margin of discretion in complex economic and technical appraisals and the limited role of review of the Court in “checking whether the relevant rules of procedure and on stating reasons have been complied with, whether the facts have been accurately stated and whether there has been any manifest error of assessment or a misuse of power”.

Lowering of thresholds to establish an abuse: following the logic of the framework described above, the Court’s treatment of the legal standards applicable to the specific abuses at issue in this case may effectively have lowered the thresholds required to establish an abuse. In the context of the interoperability case, the four-fold IMS standard was applied in such a way that, once it was established that the upstream IPR was indispensable to downstream operations, little more needed to be proved to justify a compulsory licence unless a contrary objective justification could be established by the dominant firm: in particular, the content of the new product test was starkly reduced from its inception in Magill (where the Court referred to a new product for which there was established consumer demand) to Microsoft (where it was sufficient to believe that consumer

Issue 45 ⏐ 1 November 2007 16

demand and the need to design around Microsoft’s IPRs would ensure that rivals’ implementations constituted new products). As far as tying is concerned, the judgment follows in the footsteps of earlier cases, although the Court’s interpretation of the separate product test raises the question to what extent in future essential facility cases, such as Bronner, may be caught by the tying prohibition.

What changes for firms with significant market shares as a result of the Microsoft judgment? In terms of the law, the answer is: not much; the effect may be rather that the prospect of change that they had anticipated has been set back. They will continue to struggle to find safe harbours for unilateral behaviour in the light of sweeping prohibitions derived from extreme cases. In particular, firms with market shares in excess of 80% face the threat that any exploitation of a commercial advantage that defends or enhances their market position could be regarded as abusive. Indeed, Commissioner Kroes stated in the aftermath of the Microsoft judgment that she would like to see “a significant drop in Microsoft’s market share” in PC operating systems and that such a drop would be a way of measuring the success of her antitrust policy.

Implications for EU competition policy

Modernisation: the impact of the Microsoft judgment on the Commission’s plans for modernisation of the rules for dominant firms is hard to predict. It may be tempting for the Commission not to pursue any longer a broad restatement of Article 82’s scope, given that the Court has recently endorsed per se rules in three of the four broad areas of abuse of dominance (predatory pricing, rebates/discounts and tying) and has expanded the scope of the rules in the fourth area (refusal to supply).

On the other hand, the underlying policy reasons for reform are as valid as ever, and probably even more so after the Microsoft judgment. The policy regarding dominant firms is an anachronism of the early years of EC competition law, at odds with modern economic thinking. It is inconsistent with the remainder of EC competition policy, in terms of policy goals and policy approach. The legal structure does not work because it relies upon broad sweeping prohibitions, which capture many efficient business practices, without either a proper assessment of competitive harm or the counterbalance of a well-developed theory of objective justification. As a result, the law has become more unpredictable.

The Commission may feel that a more sophisticated exercise of their prosecutorial discretion will provide a satisfactory cure for these ills. Commissioner Kroes was at pains to stress that she was committed to a “stable, fair and restrained” regulatory environment. Welcome though that is, regrettably, it does not provide a sufficient substitute for a clearly-stated and well-founded policy for Article 82. Aside from the unavoidable vagaries of enforcement policy, the Commission no longer has a monopoly over the

Technology, Media & Telecommunications

application of Article 82. National competition authorities and courts have parallel powers under either Article 82 itself or their national analogues. It is important for those jurisdictions as well as the proper administration of the Commission’s own jurisdiction that these issues should be addressed.

US/EU divergence: the current line of abuse of dominance cases seems to have set the EU institutions on a direct collision course with the US regulators. Thomas O’Barnett, the head of the antitrust division at the Department of Justice, took the view that the judgment “may have the unfortunate consequence of harming consumers by chilling innovation and discouraging competition” which in turn triggered a strong rebuke from Commissioner Kroes who said that it was unacceptable that a representative of the US judiciary should criticise a court of law outside his jurisdiction. As the European Commission becomes increasingly the global forum of choice against dominance firms, often US companies (Qualcomm, Intel and Apple are currently all in the dock), the tensions between the EU and US could well increase.

Timing: It took almost nine years from the time Sun Microsystems complained to the European Commission on 10 December 1998 to the time the CFI issued its judgment on 17 September 2007: aeons in the fast moving world of IT. At the time of Sun’s complaint, Steve Job’s iPod, which would transform the music industry, had not yet seen the light of day and nor had anyone heard of YouTube or Facebook. The industry has moved on over the last 10 years, oblivious to the antitrust battle before the European institutions, and the markets seem to have found their own solutions to the competition concerns: new web-based applications are likely to do more for competition than compulsory licensing of interface information ever will and a range of new and innovative media players such as Adobe’s Flash Player have entered the market without the help of Windows. If antitrust authorities want to play any meaningful role in high-tech industries, the administrative and judicial process has to move significantly faster.

Conclusions

Overall, the Microsoft judgment is unlikely to be the last word on the rules regarding dominant firms. The current status of the law is not tenable. The Court, which has been a driving force in modernising other areas of competition law, such as the law on restrictive agreements and EU merger control, has missed an important opportunity to bring Article 82 into the 21st century.

By Bill Allan and Christian Ahlborn, London

Issue 45 ⏐ 1 November 2007 18

Belgium

Court Orders ISP to Take Anti-P2P Action

On 29 June 2007, a new milestone was set in the legal battle over peer-to-peer (“P2P”) file sharing. In a highly controversial ruling, the Brussels Court of First Instance ordered an internet service provider (“ISP”) to implement technical measures restricting its customer’s use of P2P software.

The decision follows an earlier decision, given by the same court in November 2004, which established that copyright infringements were taking place. However, that decision did not immediately impose any measures and instead the judge ordered an expert to draft a report on whether restricting P2P traffic is technically feasible.

Origins of the Action

The Belgian Society of Authors, Composers and Publishers (“SABAM”) launched proceedings against the Belgian ISP, Scarlet (formerly Tiscali), alleging that it was knowingly permitting P2P file sharing on its network.

The aim of the action was to obtain an order against Scarlet, as a third party intermediary, to take proactive measures preventing the use of P2P networks for committing copyright infringements. The law of 30 June 1994 relating to copyright and neighbouring rights (“LAR”) allows a judge to impose such measures suspending activities infringing copyrights in summary proceeding.

First Judgment - Finding of Copyright Infringements

In the action, Scarlet defended itself against SABAM’s claim by first contesting whether it was possible to make such an order against a third party. The judge, however, agreed with SABAM on the basis that that Copyright in the Information Society Directive (2001/29/EC) explicitly provides for such a cause of action. While this Directive had not been implemented under Belgian law at that time, it had acquired direct effect as the date for implementation had passed.

Secondly, Scarlet argued that even if an order could be issued against a third party, it would not be possible to do so against Scarlet unless it had committed a wrong. As an ISP, Scarlet relied on the Belgium implementation of the E-Commerce Directive (2000/31/EC) - the law of 11 March 2003 relating to certain legal aspects of information society services (“eCommerce Act”). However, the court decided that this immunity only protects ISPs from liability for the content of transmissions; it does not prevent a finding of copyright infringements or the issuance of an order aimed at ending them.

Technology, Media & Telecommunications

Thirdly, Scarlet alleged that SABAM had not demonstrated that its customers were actually infringing copyright. This proved to be an unsustainable argument in light of the use of P2P file sharing, and was rejected.

In light of these findings, the judge considered whether to impose measures to terminate these infringements. However, as this is primarily a technical issue he designated an expert to determine how this goal could be achieved.

Second Judgment - Imposition of Remediation Measures on Scarlet

According to the expert’s report, only one technical solution was available to filter P2P traffic, though he considered the cost of applying it to Scarlet’s network to be quite high and doubted the durability of such a solution.

However, in light of the earlier decision in 2004, the court made a further judgment on 29 June 2007 ordering Scarlet to apply these measures. This was on the basis that various “Internet giants” are already using this solution and the cost was not considered prohibitive (around €0.5 per month per customer).

The court came to this decision despite the argument that it would be an infringement of the eCommerce Act, which states no general obligation can be imposed on an ISP to monitor transmitted data. The judge dismissed this argument on the basis that the E-Commerce Directive should not prevent an obligation to implement “technical surveillance instruments”. The judge also dismissed the argument that this would be a prohibited use of personal data by Scarlet, as Scarlet’s customer contracts to provide a sufficient basis for such processing of personal data.

An Appeal is Lodged

Scarlet appealed against this ruling on 6 September 2007 on the basis that it would force it to infringe the right of privacy of its customers. Scarlet also highlighted that it is no longer appropriate to treat all P2P file sharing as illegal when more and more legal content providers distribute via this technology.

Since it obtained judgment against Scarlet, SABAM has been threatening other ISPs with legal action. Over the summer, it contacted a number of them in order to try to agree a negotiated solution, but on the 12 September 2007 the Belgian ISPs Association ruled out any such compromise. Consequently, this judgment will continue to be the leading case for future developments over the use of P2P technology. Clearly the outcome of this case will be decisive for the landscape of internet service provision in Belgium, and possibly abroad.

By Guillaume Couneson, Brussels

Issue 45 ⏐ 1 November 2007 20

United Kingdom

Government Responds to Report on Personal Internet Security

The internet is an increasingly important part of our society underpinning not only global economic activity but also cultural and social behaviour. However, the growth of this online environment raises a number of risks for individuals such as becoming the victim of internet fraud, identity theft or loss of reputation. These risks can be particularly difficult to manage given the technological and jurisdictional hurdles to law enforcement on the internet.

In light of these risks, the House of Lords Science and Technology Committee launched an inquiry into safety on the internet and published a report entitled Personal Internet Security in August 2007. The report contained a number of wide reaching proposals to improve internet security by amending the law, raising public awareness and restructuring enforcement activity.

The Government has now responded to the proposals in the report. While agreeing that the welfare of users of the internet is increasingly important, it does not believe that the public has lost confidence in using the internet and is not intending to adopt all of the recommendations in the report.

Two particular areas of interest are the proposals for a notice of breach law and the removal of the “mere conduit” exemption.

Notice of Breach

One of the key proposals made in the report was that a data security breach notification law should be introduced in the United Kingdom - i.e. a law requiring businesses to tell either its regulators or their customers if there is an unauthorised disclosure of their customers’ personal information. This type of law would not only provide greater visibility about the level of threat posed by data breaches but would also provide a incentive for businesses for ensure they keep their customers’ data secure.

The Government’s response acknowledges that the introduction of this type of law in other jurisdictions (most notably the US) is an “interesting development”. However, the Government is less convinced that it would lead to an immediate improvement in the protection of personal information. A number of factors lead to this conclusion:

• the introduction of similar laws in other jurisdictions may have desensitised consumers to security issues and undermined confidence in the internet;

• other jurisdictions, such as the US, do not have general data protection laws so there is a greater need for this type of market lead protection; and

Technology, Media & Telecommunications

• in certain circumstances notifying consumers of the security breach might not be appropriate (for example, it might highlight the value of the data that has been lost).

While this issue remains on the Government’s watch list, and European proposals to introduce this requirement in the communications sector remain, there is no indication that the Government will introduce a notice of breach law of its own volition in the immediate future.

Removal of the “Mere Conduit” Exemption for ISPs

The report also recommended that the “mere conduit” exemption should be removed once an ISP has detected or been notified that a machine on their network is sending out spam or viruses. The removal of this exemption could potentially make ISPs liable for any damage to third parties resulting from a failure immediately to isolate the affected machine.

This suggestion was also rejected by the Government, mainly because it arises out of the E-Commerce Directive (2000/31/EC) and therefore must be implemented by the United Kingdom. Moreover, from a practical perspective, an ISP will want to take action against compromised machines in any event to protect its own network and normally include this right in their terms and conditions.

The report is available here and the Government’s response is available here.

By Peter Church, London

Issue 45 ⏐ 1 November 2007 22

Media & Broadcasting

Czech Republic

New Digital Television Laws Proposed

Experimental digital terrestrial television (“DTT”) was launched in the Czech Republic back in May 2000. Seven years later, the move from analogue to digital still lacks momentum, the rules for the licensing and operation of DTT are confused and the network infrastructure remains underdeveloped.

Despite strong interest in DTT among broadcasters and consumers alike, the digital boom is being held back by an inadequate legal framework. This is expected to change with the introduction of new legislation to improve legal certainty and confidence in the sector by setting out clear and transparent rules of play.

The “Digital Packet”

In September 2007, the Chamber of Deputies of the Czech Republic passed the long-awaited “packet” of amendments to Czech media law to permit a full launch of DTT in the Czech Republic.

The core of the so-called “Digital Packet” amends the two laws setting out the framework for digital broadcasting:

• the Electronic Communications Act (Act No. 127/2005 Coll., on Electronic Communications and Amendment to Certain Acts); and

• the Broadcasting Act (Act No. 231/2001 Coll., on Provision of Radio and Television Broadcasting and Amendment to Certain Acts).

Licensing Procedure for Broadcasters

Prospective DTT broadcasters will no longer need to compete for licences in an open tender. Licences for DTT broadcasting now will be issued to any applicant who can meet a set of qualifying conditions. However, DTT broadcasting based on licences acquired under these relaxed conditions will be possible only after the digital switchover, which will take place on 10 October 2010 at the earliest and could be delayed until 31 December 2012.

The licence itself will, nevertheless, no longer automatically guarantee the allocation of a particular radio frequency. Instead, it will be the responsibility of the broadcaster to team up with one of the DTT network operators. Following the licence award, a DTT channel must be launched within 360 days or the licence will expire.

Regulation of Network Operators

DTT network operators will be required to publish a reference offer. This offer must be public (e.g. available on their websites) and set out detailed

Technology, Media & Telecommunications

terms on which they are prepared to provide services to DTT broadcasters. The reference offer, and any subsequent changes, must become effective not earlier than one month following their publication. DTT network operators will also have to publicise information about any free capacity in their network and the scope and structure of the network capacity that is in use.

The Digital Package establishes a set of minimum requirements for DTT networks. They must enable the broadcast of at least four channels in PAL standard (720x576 resolution) or better. Since this requirement applies to all types of network, some fear that it will hinder the launch of digital TV on mobile phones, where a lower 352x288 resolution is more appropriate.

The Digital Package also requires that network operators conclude a contract for the provision of services with every DTT broadcaster who demands contractual terms: (i) that comply with the reference offer for that network operator; (ii) that correspond to the technical parameters of the network; (iii) that will not damage the integrity of the network; and (iv) that offer a price equal to or greater than the “cost-oriented” price. This is, of course, subject to the network operator having sufficient capacity on its network to carry that channel.

In the event that negotiations between the network operator and broadcaster break down, the Czech Telecommunication Office can step in to resolve the dispute at the request of either of the parties involved.

Original Broadcasters Given Priority over Retransmission Broadcasters

Retransmission broadcasters (i.e. broadcasters who do not create their own material but instead rebroadcast other channels or substantial parts of other channels) will no longer be limited to cable and satellite networks and will also be able to use DTT.

However, for public policy reasons, their access to DTT will be limited in order to promote original broadcasters (i.e. those who create new material themselves). Retransmission broadcasters will be obliged to surrender their DTT transmission capacity to any original broadcasters who have concluded a contract for DTT services with the same network operator. The retransmission broadcasters must surrender their transmission capacity within 90 days of the receipt of notice from the original broadcaster, otherwise the retransmission broadcaster’s registration may be terminated eventually.

Must-Carry Obligation for Cable and Mobile

Network operators providing DTT broadcasting services for mobile phones and retransmission broadcasters operating on cable networks will be required to include, as a part of their minimum subscriber package, all free-to-air national DTT channels. This obligation will, however, expire after the digital switchover.

This part of the Digital Package was particularly controversial. Some argue that this requirement is too onerous and unduly restricts the commercial

Issue 45 ⏐ 1 November 2007 24

freedom of the cable and mobile operators. On the other hand, the must-carry obligation is only temporary and, as a consequence, it will ensure a level playing field for all broadcasters, who will all get access to the same share of the audience.

Compensatory Licences for Existing Broadcasters

When the new law comes into force, the two commercial broadcasters (Nova and Prima), as well as the public service broadcaster (Czech Television) with its four channels, will have their licences reconfirmed. Nova and Prima will also have their licences prolonged by eight years, until 2025 and 2026 respectively.

On top of that, Nova and Prima will be able to receive an additional compensatory broadcast licence up until digital switchover provided they:

• comply with the digital switchover date announced by the government; and

• hand back their analogue licences in accordance with the schedule to be announced by the government.

Therefore the compensatory licences are both an incentive for Nova and Prima to comply with the digital switchover conditions and also a compensation for their voluntary surrender of their analogue licences, which are necessary to free-up spectrum for the digital switchover.

Six other broadcasters can also claim compensatory licences: Febio TV, Ocko, RTA, TV Barrandov, TV Pohoda and Z1. These broadcasters have also formerly received “standard” licences for DTT broadcasting, though the licence award has been appealed. The Digital Packet provides for the termination of these pending licensing proceedings. As a temporary measure, the six broadcasters will receive the compensatory licences which will enable them to launch DTT broadcasts in the same network as originally anticipated under their “appealed” licences. Their compensatory licences will also expire after the digital switchover. As regards the period after the digital switchover, the six broadcasters will be able to apply for regular DTT licences under the new licensing procedure introduced by the Digital Packet.

The final reason for the award of the compensatory licences is to improve the content available on DTT. Currently, there are not enough attractive programmes to persuade the television viewers to switch to DTT. Therefore, the short-term grant of extra compensatory licences should make the programme choice richer and more appealing.

After the digital switchover the compensatory licences should no longer be needed as there will be sufficient competition among new broadcasters to drive up the standard of the channels.

Final Approval by the Senate

Recently, the Digital Packet has been also approved by the Senate, which gave it the green light without any further ado. If ratification by the President

Technology, Media & Telecommunications

is also forthcoming, the Digital Packet could enter into force as early as 1 January 2008.

However, this is not the last step on the road to digitalisation by any means. The Digital Packet will be followed soon by additional technical amendments which are necessary for the adjustment of the legal framework in order that DTT can be fully rolled out.

By Zuzana Viktorinová, Prague

Germany

A Split Licence Award for DVB-H Mobile Television

The German telecommunications and broadcasting regulators have recently awarded spectrum and broadcasting licences to pave the way for the roll-out of DVB-H (Digital Video Broadcasting - Handheld) mobile television in Germany. It is estimated that this service should be launched in the first urban areas before the European Football Championship in 2008.

Federal and State Licensing

In Germany, there is a constitutional separation between telecommunications law (which is regulated at federal level) and media law (which is regulated at state level). This means that the tender for the necessary authorisations for DVB-H had to be carried out by both the Federal Network Agency and the State Media Authorities.

The Federal Network Agency awarded the rights to use the relevant DVB-H frequencies for the actual transmission of mobile television whereas the State Media Authorities awarded the broadcast licences entitling the owner to operate a programme platform for mobile TV (i.e., the right to aggregate and bundle content and to transmit that content over the frequencies awarded by the Federal Network Agency).

Spectrum Award

On 8 October 2007, the Federal Network Agency awarded the frequency rights for the transmission of DVB-H signals in Germany to T-Systems Media & Broadcast, a subsidiary of Deutsche Telekom AG.

T-Systems Media & Broadcast already has almost exclusive rights over the DVB-T (Digital Video Broadcasting - Terrestrial) network in Germany and also provided the transmission network for the first DVB-H trials in Germany during the Fifa World Cup 2006. Regardless of the fact that this could lead to limited competition in the digital television market, the regulator decided that T-Systems Media & Broadcast demonstrated the best technical capability in its bid and also guaranteed greater geographical coverage.

Broadcast Licence Award

Issue 45 ⏐ 1 November 2007 26

The State Media Authorities followed suit on 16 October 2007, indicating that they intended to grant the broadcasting license for the operation of the DVB-H platform to Mobile 3.0, a joint venture between Mobiles Fernsehen Deutschland (“MFD”) and Neva Media.

Neva Media is backed by publishing houses Hubert Burda Media and Holtzbrinck as well as media managers Paulus Neef, Bernd Curanz and Christiane zu Salm, among others. The biggest single shareholder of MFD is South African media and technology company Naspers, which has already established DVB-H broadcasting networks in several African countries. MFD also operates the first German mobile TV platform, known as “watcha”, which is provided using DMB (Digital Multimedia Broadcasting). This was launched in June 2006 and is currently available in 16 urban areas and distributed by mobile service providers.

According to the State Media Authorities, Mobile 3.0 now has until 9 November 2007 to present a plan for the allocation of the available capacity to television broadcasters, including a regional offer, as well as for the allocation of the four available radio slots.

If Mobile 3.0 is then granted the licence, several TV channels and other content providers will have to make arrangements with Mobile 3.0 as the content platform operator.

Commentary

The proposed award of the broadcasting licence for the DVB-H platform to Mobile 3.0 is a set-back for mobile operators T-Mobile, Vodafone and O2 who set up a joint venture for the provision of mobile TV earlier this year and jointly bid for the broadcasting licence.

It is likely that Mobile 3.0 and T-Systems Media & Broadcast will now have to enter into negotiations over how to actually set up a transmission infrastructure for mobile television in Germany. Mobile 3.0 will also have to enter into negotiations with the mobile network operators and mobile service providers in Germany in order to establish a distribution network and obtain access to end customers.

By Christoph Enaux and Karen Sokoll, Berlin

The Return of Media Mogul Leo Kirch

The Munich media entrepreneur, Leo Kirch, will be in charge of licensing the national television rights for the football Bundesliga for the period 2009 to 2015. These football rights will be tendered early next year by the German Football League (“DFL”).

Auction of Rights

DFL has appointed Sirius, a newly established agency and subsidiary of Leo Kirch's KF 15 GmbH, to organise the auction of these rights for the next two licensing periods starting in spring 2008. Each of these licence periods will run for three years. Sirius has guaranteed DFL a minimum revenue of €3

Technology, Media & Telecommunications

Berlin Linklaters LLP Rankestraße 21 10789 Berlin Postfach 30 18 50 10746 Berlin Tel: (+49) 30 21496-0 Fax: (+49) 30 21496-100 Brussels Linklaters LLP Rue Brederode 13 B - 1000 Brussels Tel: (+32) 2 501 94 11 Fax: (+32) 2 501 94 94 London Linklaters LLP One Silk Street London EC2Y 8HQ Tel: (+44) 20 7456 2000 Fax: (+44) 20 7456 2222 Paris Linklaters LLP 25 rue de Marignan 75008 Paris Tel: (+33) 1 56 43 56 43 Fax: (+33) 1 43 59 41 96 Prague Linklaters, v.o.s., advokátní kancelář Palác Myslbek Na Příkopě 19 117 19 Prague 1 Tel: (+420) 221 622 111 Fax: (+420) 221 622 199

28

This publication is intended merely to highlight issues and not to be comprehensive, nor to provide legal advice. Should you have any questions on issues reported here or on other areas of law, please contact one of your regular contacts, or contact the editors.

© Linklaters LLP. All Rights reserved 2007

Please refer to www.linklaters.com/regulation for important information on our regulatory position.

We currently hold your contact details, which we use to send you newsletters such as this and for other marketing and business communications.

We use your contact details for our own internal purposes only. This information is available to our offices worldwide and to those of our associated firms.

If any of your details are incorrect or have recently changed, or if you no longer wish to receive this newsletter or other marketing communications, please let us know by emailing us at marketing.database@linklaters.com

Linklaters converted to Linklaters LLP on 1 May 2007. References in this document to Linklaters for the period following 1 May 2007 accordingly refer to Linklaters LLP and, where relevant, its affiliated firms and entities around the world.

Linklaters LLP is a limited liability partnership registered in England and Wales with registered number OC326345. The term partner in relation to Linklaters LLP is used to refer to a member of Linklaters LLP or an employee or consultant of Linklaters LLP or any of its affiliated firms or entities with equivalent standing and qualifications. A list of the names of the members of Linklaters LLP together with a list of those non-members who are designated as partners and their professional qualifications is open to inspection at its registered office, One Silk Street, London EC2Y 8HQ or on www.linklaters.com and such persons are either solicitors, registered foreign lawyers or European lawyers

Peter Church Email: peter.church@linklaters.com

/ /

Editor:

billion for this period, approximately 20 per cent more than the football clubs receive at present.

For the international licensing of the rights, DFL will establish a foreign subsidiary for the first time. DFL expects to generate revenues of approximately €460 million for the period from 2009 to 2015.

Production Joint Venture

Moreover, DFL has set up a joint venture production company between Sirius and DFL, in which Sirius will hold 51% of the shares. This joint venture will be responsible for the production of television footage of the football games (including pay-TV live transmission). All pay-TV providers will therefore be provided with the same pre-produced television footage.

This move has been criticised by some members of DFL as well as various television broadcasters. However, it has been introduced to ensure that smaller cable operators or other broadcasters will be able to bid for the necessary rights without having to invest large amounts in the production of the footage.

Return of the Mogul

The new agreement with Kirch is a major development for the German sports rights market. Kirch was involved in one of the greatest failures in the history of the German economy when his media group became insolvent in 2002. Before the insolvency, he was one of the most important players in the German sports and television market. The Kirch Group included, amongst others, the pay-TV provider Premiere as well as the TV channels that now belong to ProSiebenSat.1 Media AG.

His return is now likely to make it more expensive for Premiere to win back the pay-TV rights for the football league.

By Christoph Enaux and Karen Sokoll, Berlin