ECJ ruling encourages copyright holders to shoot the messenger

In their decision issued last week in the case of UPC Telekabel v Constantin Film, the European Court of Justice (ECJ) confirmed that EU states have the right to issue injunctions requiring internet service providers (ISPs) to block their internet users’ access to copyright-infringing websites.

Whilst at first glance the decision may appear to be a landslide victory for rights owners, the decision also offers welcomed clarity to ISPs on the extent of their obligations to prevent infringements.

The ECJ’s decision arose out of an Austrian case involving an application by two film producers, for an injunction requiring an Austrian ISP to prevent its users accessing a website that enabled its users to stream or download films which infringed the producers’ copyright.

According to Article 8(3) of the Copyright in the Information Society Directive (2001/29) (Directive), national courts have the power to grant a blocking injunction against an ‘intermediary’ whose services are used by a third party, to infringe copyright or related rights. The ECJ’s decision addresses two key issues; firstly, what sort of intermediaries can be subject to injunctions under the Directive, and secondly, what form of injunctions can be granted against them?

Can an injunction be granted against an ISP as an intermediary?

The primary issue to be determined by the court was whether a party making infringing information available on their website ‘uses’ the services of the ISP whose customers access the website to do so, where the ISP’s customers themselves have not committed an infringement, thus providing a jurisdictional platform for courts to grant injunctions against such an ISP.

UPC argued that they had no contractual business relationship with the website operators, having neither made internet access or storage space available to them. Since it could not be established that their direct customers acted unlawfully (although, as the Advocate General Cruz Villalon pointed out in his preceding opinion delivered back in November, it could be assumed with near certainty that they took advantage of the services on offer on the offending website), they maintained that it could not be considered that their services had been used to infringe a copyright or related rights. UPC also emphasised that, in any event, the various blocking measures available were excessively costly, especially as for the most part they could be easily circumvented with minimal technical expertise.

Contrary to the assertions of UPC, the ECJ found that where an ISP’s service users access infringing content on a website, that ISP is in fact an intermediary whose services are used to infringe copyright within the scope of the Directive, and as a result they can be subject to injunctions forcing them to block access to the offending sites. They cited the rationale that the ISP “is an inevitable actor in any transmission of an infringement over the internet between one of its customers and a third party, since in granting access to the network it makes the transmission possible.”

In their decision, the ECJ recognised that, on a practical level, as the services of intermediaries are increasingly used to infringe copyrights, such intermediaries are often ideally placed to take preventative action. The ECJ emphasised that in order to fulfil the key objective of the Directive, which is to guarantee right holders a high level of protection as outlined at Recital 9 of the Directive, ISPs must be included within the parameters of Article 8(3), because to rule otherwise would substantially diminish the protection afforded to such right holders.

This means that in future it will not be necessary to show a specific relationship between a person infringing copyright and the intermediary against whom an injunction may be issued, nor will it be necessary to prove that the customers of the ISP in question actually access the protected subject matter on the known infringing website, as the ECJ reiterated that that spirit of the Directive required Member States to not only take action to bring existing infringements to an end, but also to prevent further infringements, or at least make them more difficult to commit.

Can an injunction be granted without specifying the means of implementing it?

Having decided that an injunction could be granted against an ISP in the circumstances outlined, the ECJ then proceeded to consider the nature of the injunction to be granted, and in particular whether the Austrian court’s approach in leaving the ISP to decide the means to be used in blocking the website was acceptable, or whether the court should be required to specify any prescribed block should be implemented.

In reaching their decision, the ECJ recognised the tension between the ISP’s freedom to conduct a business, the internet users’ freedom of information, and the right holders’ copyright, and emphasised the need to achieve a “fair balance” between these fundamental rights. In assessing whether such a balance had been struck, the ECJ found that by giving the ISP a wide discretion to determine the appropriate measures to implement a block of the website, the Austrian Court’s ‘results’ form of injunction had in fact ensured the optimum business freedom of the ISP, as they could choose to put in place measures best adapted to their available resources and the challenges of carrying on their particular business activities. Furthermore, provided the ISP had taken all measures “capable of being considered reasonable”, they could rest assured that they would not be held liable for breach of the injunction, thus ensuring that an ISP cannot be expected to make unbearable sacrifices in order to protect the conflicting interests of a rights-holder.

In this regard the ECJ judgment differed from the opinion of the Advocate General, where he expressed the view that it would be “incompatible with the weighing of the fundamental rights of the parties to prohibit an ISP generally [from accessing an infringing website] and without ordering specific measures”.

The Advocate General did, however, acknowledge that a specific blocking measure imposed on an ISP relating to a specific website would not automatically be disproportionate simply because it entailed not inconsiderable costs and could be easily circumvented without any special technical knowledge.

The ECJ imposed two conditions on the granting of a general injunction such as that proposed by the Austrian court in order to ensure a fair balance is struck between the fundamental rights of the parties:

(i)            Firstly, measures must not unnecessarily deprive users of the possibility of lawfully accessing the information available, so in other words measures must be strictly targeted to ensure that internet users’ right to freedom of information is not be curtailed more than is necessary; and

(ii)           Secondly, those measures must have the effect of preventing unauthorised access to the protected subject-matter or, at least, of making it difficult to achieve and of seriously discouraging users accessing the infringing subject-matter. This means that courts will not decline to grant an injunction just because there is no fool-proof solution available. As the English High Court has acknowledged “a blocking order may be justified even if it only prevents access by a minority of users”.

The ECJ declined to elaborate further on how best to balance and protect the competing rights of parties in such a case, leaving the ultimate decision as to “fair balance” in the domain of national courts.

In order to ensure that the fundamental rights of internet users are afforded adequate protection, and are not diminished in the wake of an insurgence of applications for injunctions by rights-holders, the ECJ also introduced the concept that internet users can assert their rights before national courts, ensuring that they have a forum for redress where they believe measures imposed by an ISP are unduly restrictive.

The battle is won, but the war has just begun…

Whilst the International Federation of the Phonographic Industry (IFPI) has issued a positive statement in support of the ECJ ruling, which undoubtedly represents a string to their bow in fighting online piracy, the decision is unlikely to have a notable impact on the approach of the UK courts to website blocking. The High Court had already been persuaded to grant specific injunctions against a number of ISPs under domestic intellectual property legislation before the recent judgment came to light, and is likely to continue to grant such orders without hesitation in the future, particularly with the reinforcement of an ECJ decision behind them. Other EU member states are now likely to follow suit, adopting an approach more consistent with that already prescribed by English law.

Whilst those campaigning for better protection of intellectual property rights are celebrating their apparent victory, they may stop to consider that whilst in principle the ECJ ruling is a step in the right direction, in practice the operators of illegal websites, and the ISPs making them available online, are often based outside of Europe or conceal their identity, which means that in reality it is very difficult to pursue them before the courts. Let the historic examples of the difficulty encountered in shutting down notorious copyright-infringing sites such as Pirate Bay serve as a cautionary warning not to celebrate too soon – the road ahead for rights-holder’s crusading against piracy in the internet age is a treacherous one; and for every website you successfully shut down you can all but guarantee the same will resurface under a different address or hosting provider!

Twitter launch ‘report tweet’ function

Twitter has introduced a new ‘report tweet’ function across all of its platforms to enable users to report abusive tweets.

The move has been hailed by campaigners who called for social network websites to take action against threatening and abusive users. As noted in our earlier blog post, ‘report abuse’ buttons and similar functions can have limited effect in stopping internet trolls. But this is a step forward in terms of social network websites taking some responsibility for preventing their platforms being used for abusive purposes.

‘Anything you tweet may be given in evidence’

There are stories in the news today warning social network users that their tweets and other messages could be used as evidence against them in court cases.

The issue came to the attention of the press after US teenager, Cody Hall, had a manslaughter charge upgraded to murder as a result of boasts about speeding he put on Twitter months earlier. The tweets including “Live fast die young” and “come on a death ride with me” lead prosecutors in California to increase the charge against him from manslaughter to murder and to revoke his bail following his arrest for knocking over and killing cyclist Diana Hersevoort whilst driving in Dublin, California.

Whilst the story might be surprising to some, there has not been a recent change in the law. It has always been the case in the UK that things said on social media can be seen by the police and Crown Prosecution Service (CPS) and used in court.

Earlier this year, the police passed a file to the CPS for investigation after tweets by trainee accountant Emma Way stating “Definitely knocked a cyclist off his bike earlier. I have right of way – he doesn’t even pay road tax!” were forwarded to the police by other Twitter users. It also seems likely the Facebook profile of Francisco Jose Garzon Amo, the driver of the speeding train that derailed in Spain last month killing 79 people, will be used as evidence against him after he allegedly posted a picture of a train speedometer at 200km/h (124mph), writing: “I’m at the limit and I can’t go any faster or they will give me a fine.”

It is not just post about crimes which social network users should be concerned about, earlier this year a tweet published by Sally Bercow about Tory peer Lord McAlpine was found by the High Court to be libellous. Mrs Bercow, the wife of Commons Speaker John Bercow tweeted “Why is Lord McAlpine trending *innocent face*”, two days after BBC Newsnight wrongly linked a “leading Conservative politician” to sex abuse claims. Following the ruling, a damages settlement was agreed for an undisclosed amount.

The message for social network users is clear; do not think that because you say something on Twitter or Facebook that it cannot be used against you in court. Remember that unlike a private conversation with a friend, posts on social networks can be viewed almost instantly by people all around the world. Only post comments which you are comfortable sharing publically with others.

What can be done to stop internet trolls?

With recent headlines about feminist Caroline Criado-Perez being bombarded with violent and misogynistic tweets and 14-year old Hannah Smith who committed suicide after being taunted on social networking site Ask. fm, questions about being asked about what can be done to stop internet trolls?

Twitter, Facebook and other social networking sites have rules for using their service which are signed up to by users as part of the account registration process. The sites also have processes in place to enable users to report abuse either via a report abuse button or an online form. However, even where abuse is reported, the ability of the social networking sites to put a stop such abuse is limited. Their key sanction is usually to suspend the user’s account but determined trolls can often run multiple accounts at one time or will open another fake account almost immediately after their access is suspended.

Twitter’s Abusive Behaviour Policy encourages users to share their views but discourages targeted abuse and harassment stating that “Twitter is a platform that provides a global communication service which encompasses a variety of users with different voices, ideas and perspectives. As a policy, we do not mediate content or intervene in disputes between users. However, targeted abuse or harassment may constitute a violation of the Twitter Rules”. The policy goes on to say that “Users are allowed to post content, including potentially inflammatory content, provided they do not violate the Twitter Terms of Service and Rules.”

The Twitter Rules expressly forbid certain posts stating:

  • “You may not publish or post direct, specific threats of violence against others” and
  • “You may not use our service for any unlawful purposes or in furtherance of illegal activities. International users agree to comply with all local laws regarding online conduct and acceptable content.”

In certain circumstances, violent or threating tweets can also be brought to the attention of the police. However, there can be practical difficulties in finding trolls as they can often access social networks using a number of different fake accounts and public computers. Even when trolls can be identified, the police have limited powers to prosecute. The key offence which can be committed by trolls is under the Communications Act 2003:

  • Section 127(1)(a) of the Communications Act 2003 provides that a person is guilty of an offence if he sends by means of a public electronic communications network a message or other matter that is grossly offensive or of an indecent, obscene or menacing character.
  • Section 127(3) of the Communications Act provides that a person guilty of an offence under section 127 is liable, on summary conviction, to imprisonment for a term not exceeding six months or to a fine not exceeding level 5 on the standard scale or to both.

Keir Starmer QC the Director of Public Prosecutions recently published guidelines for prosecutors who are taking on cases involving communications sent via Twitter and other social networks. The guidelines provide that prosecutions should be sought where there is a credible threat of violence, a targeted campaign of harassment against an individual or a breach of court orders, but provide that a high threshold must be applied whether deciding whether communications are grossly offensive.

The guidelines call for prosecutors to recognise the right to freedom of expression and provide that prosecution should only be brought when the communication is “more than offensive, shocking or disturbing, even if distasteful or painful to those subjected to it”.

In addition to the steps taken by the police and social networking sites to stop internet trolls, employers and schools could also help to limit trolling by making the use of school/work computers for offensive purposes a breach of the school/employment policies which could lead to suspension or sacking.  Education can also play a part and peer pressure from other social network users could help to reduce the prevalence of trolling. Unfortunately, without better mechanisms for identifying internet trolls and stricter sanctions for those who are caught, it seems likely that trolling is set to continue for the foreseeable future.  

ASA uses its power to ban a Twitter campaign for the first time

Speaking at the Cannes Lions Festival of Creativity on 19 June, Coca Cola’s most senior marketer Joseph Tripodi called on marketers to take a “leap of faith” and embrace social media as a brand building tool. However, as Nike discovered the very next day, advertising using social media is not free from constraints.

Since 1 March 2011 the Advertising Standards Authority (ASA) has had the power to oversee businesses’ marketing communications on their own websites, as well as on social networking sites and other “non-paid-for” space online, to ensure that they comply with the CAP (Committee of Advertising Practice) Code.

The first major case that forced the ASA to look at advertising on social media came to light earlier this year when it launched an investigation into tweets by celebrities such as Katie Price and Rio Ferdinand promoting Snickers. The campaign involved celebrities posting a string of bizarre tweets ending with “You’re not you when you’re hungry@snickersUk#hungry#spon” and a picture of them holding a Snickers. The ASA ultimately dismissed the complaints against Mars finding that the inclusion of the #spon hashtag in the final “reveal tweets” made them clearly identifiable as marketing communications.

There has since been a noticeable increase in the number of sponsored tweets or “tweeting for money” and this looks set to continue. However, in the first case of its kind, the ASA has taken action to “ban” a campaign which features them. As part of its “Make it Count” campaign, Nike UK used the personal Twitter account of footballer Wayne Rooney to post the following tweet:

Nike posted a similar tweet on the account (subsequently deleted for unconnected reasons) of Arsenal footballer Jack Wilshere:

Jack Wilshere – “In 2012, I will come back for my club – and be ready for my country. #makeitcount”.

Responding to a complaint that the tweets were not clearly identified as advertising, Nike claimed that both footballers were well known for being sponsored by Nike and argued that Twitter users would not be misled about its relationship with the players. Nike took the view that the presence of the Nike URL and campaign strap line #makeitcount within the body of the tweets, indicated that the purpose of the tweets was to direct followers to the Nike website and made it sufficiently clear that the tweets were advertising.

The ASA disagreed, finding that the reference to Nike was not prominent and could be missed, making the tweets not obviously identifiable as advertising and putting them in breach of the CAP Code. The ASA held that as not all Twitter users would know about the players’ sponsorship deals with Nike, the tweets should have featured an indication hashtag, such as #ad or #spon, to make it clear that they were marketing communications.

Just the one complaint?

It is interesting to note that the Nike campaign was banned by the ASA despite only receiving one complaint. To coincide with its 50th anniversary, the ASA has recently released a list of the most complained-about ads of all time.

Top of the list was a TV advert for Kentucky Fried Chicken which aired in 2005 and featured call centre workers singing with their mouths full of food. The ad received a record 1,671 complaints with many people considering that it could encourage bad manners among children. However, despite the record number of complainers, the complaint was not upheld by the ASA, which ruled that the ad was unlikely to change children’s behaviour or undermine parental authority.

The other ads to make the top 10 were:

2. Auction World (2004): Shopping channel – 1,360 complaints – referred to Ofcom

3. Paddy Power (2010): Cat being kicked by blind football player – 1,313 complaints – not upheld

4. The Christian Party (2009): Poster saying “There definitely is a god” – 1,204 complaints – not upheld

5. British Safety Council (1995): Condom advert featuring Pope – 1,192 complaints – upheld

6. Marie Stopes International (2010): TV ad offering sexual and reproductive healthcare advice –  1,088 complaints – not upheld

7. Volkswagen (2008): Depicted an engineer fighting multiple versions of himself – 1,070 complaints – partially upheld

8. Yves St Laurent (2000): Poster of naked reclining Sophie Dahl – 948 complaints – upheld

9. Department of Energy and Climate Change (2010): Press and TV campaign about climate change – 939 complaints – upheld in part

10. Barnardo’s (2008): TV campaign about domestic child abuse – 840 complaints – not upheld.

Battle of the business models

Wired magazine has a fascinating interview with Jeff Bezos, CEO of Amazon, to coincide with the first shipping of the Kindle Fire, Amazon’s rival to the iPad.

What the interview highlights is the way the internet (at least in the English-speaking world) is increasingly concentrating into four “ecosystems” – Google, Apple, Facebook and Amazon – with each of these having a distinctive business model:

  • Apple’s model is hardware-centric. The content it sells through iTunes and the App Store is a means to an end, the end being to sell its highly profitable, premium-priced devices such as the iPhone, iPad and Mac. Its lower-priced devices such as Apple TV and the iPod Touch serve the same end, operating (as I can testify from personal experience!) as a “gateway drug” to the more expensive models by ensuring people buy into the ecosystem.
  • Amazon’s model is content-centric. This is the polar opposite to Apple: Amazon makes money from selling content, and it therefore keeps its device prices at rock-bottom in order to draw people into its content ecosystem. In his interview with Wired, Bezos doesn’t rule out literally giving away the Kindle in future.
  • Google’s model is data-centric. Its mission statement is “to organize the world’s information and make it universally accessible and useful” – not least to make it useful to its advertisers. Hence Google generally gives away its products for free, whether that’s services like Gmail and Picasa for consumers, or the Android operating system for mobile phone developers and networks. The products are aimed at encouraging people to put more and more of their information into Google’s servers.
  • Facebook’s model is social-centric (sorry!). Like Google, its aim is to collect as much data as possible about people so that it can then sell advertising. However, it comes at this at a different angle from Google, building out from people’s social relationships – highly valuable information that is donated to it by its 800m users.

It remains to be seen which of these models will win out or how they will coexist. However, a couple of questions come out of this.

First, you’ll notice that the above list makes no mention of Twitter. Twitter may be a hugely popular service, but it’s a long way from constituting an ecosystem or dominant business model to compete with Apple, Amazon, Google and Facebook. It’s something that people use, not somewhere that people live.

Second, what does this mean for smaller companies seeking to make money online? Increasingly the routes to do so lie through one or more of those ecosystems: through developing apps for Apple, Android or Facebook, through ensuring a strong presence on Google, through having content available through Amazon. This provides people with a lot of opportunities – just see how the App Store can make software available to tens of millions of potential customers – but also reflects something of a reduction in the “free for all” that has driven innovation online to date.

Something of that concern for a reduction in freewheeling innovation can be seen in Bezos’ criticism of the software and business method patents that helped Amazon in the past, but which he now clearly sees as more of a threat:

For many years, I have thought that software patents should either be eliminated or dramatically shortened. It’s impossible to measure the toll they’ve had on the software industry, but on balance, it has been negative.

Whatever your view on software patents, they are certainly playing a central role in the concentration of internet business into a small number of ecosystems, between which fierce battles rage in patent courts around the world. Again, it remains to be seen what effect this will have on innovation for smaller companies.

Death of the domain name?

Interesting article on plans for forthcoming releases of the Google Chrome and Mozilla Firefox browsers to “deemphasise” the address bar, so that the URL of the page you’re viewing is not visible in normal browsing.

As the report puts it:

Google’s motivation to reduce your dependence on the URL bar is clear, since the company would rather that you think of using the web the same way you use your iPhone or Android device.

As Google sees it, the Web isn’t a collection of sites such as, or Instead these are all software applications called The New York Times, Gmail, and Facebook that happen to live online instead of on the desktop.

From an online safety point of view, this has pros and cons. Critics will argue that hiding the URL will make life easier for scammers to direct people to phony sites. Against that, the linked report suggests that removing human error (typing the wrong address) will reduce some opportunities for scam sites.

I remember suggesting at an event some years ago that search would ultimately displace URLs/domain names as the main means by which people navigated the web. It’s taken longer than I thought, but it does now seem to be happening: I’ve noticed a number of offline advertisements in recent months (such as billboards and magazine ads) that invite people to search for a keyword rather than giving the advertiser’s website address.

Add to that this new trend to treat websites as apps, and it may well be that the URL is on its way towards becoming a purely technical feature working in the background.