Ref’s Net penalty decider

This week the recently retired Lord Justice Jacob referred an important question to the ECJ in Football Dataco v Sportradar [2011] EWCA Civ 330. It’s about whether someone who uploads infringing content on the internet is liable in the country where he is – or where the content is viewed – or both.

Football Dataco compile Football Live, a database of UK football statistics updated during matches (goals, goal scorers, yellow and red cards, substitutions). Sportradar provides a competing service, Sport Live Data, with servers in Germany and Austria, accessible in the UK via betting sites. Football Dataco claim that Sportradar are copying data from Football Live, constituting infringement of UK copyright and database right.

Last November, Floyd J, determining whether the English courts had jurisdiction, held in [2010] EWHC 2911 (Ch):

1. There was a good arguable case that Football Dataco’s content was protected by copyright, that Sportradar had copied a part of it and had authorized/was jointly liable for copyright infringements by punters in the UK who had reproduced the data on their computer screens.
2. There was a good arguable case that Sportradar would be jointly liable with punters for database infringement by extraction and with the UK betting site for re-utilization.
3. Sportradar had not itself re-utilized Football Live in the UK. ‘Re-utilization’ is defined as ‘making available’, which Floyd J concluded would be committed where Sportradar’s servers are, not in the UK.

All points were appealed and on Tuesday the Court of Appeal held:

1. Football Live might be protected by copyright but there was no copyright infringement if Sportradar had copied data from it. ‘Its recording may sometimes involve some skill (who scored in a goalmouth scramble) but it is not creative skill.’ It followed that Sportradar could not be authorizing or jointly liable for punters’ copyright infringement.
2. There was a plausible case of Sportradar being jointly liable for database infringements by the UK betting site and punters.
3. The question of whether ‘making available’ occurs where a site is hosted and/or where it is viewed required a reference to the ECJ.

It was only in December that Jacob LJ referred to the ECJ the question: ‘Does “author’s own intellectual creation” require more than significant labour and skill from the author, if so what?’ He appears to now know at least part of the answer (creative skill).

The jurisdictional question now being sent to the ECJ concerns the database right not copyright, and the copyright and database ‘making available’ rights are not identical. However a parallel question exists in relation to copyright and it would seem likely that whatever the ECJ decide for the database right would also apply to copyright. The legislative history and case law of copyright’s making available right therefore come into consideration.

Sportradar make the policy argument that if making available on the internet happens everywhere that a site can be seen, then websites have the impossible task of making themselves compliant with all the different national laws in the world. A counter policy argument would be: if liability were only wherever servers are located, then infringement is avoided by using servers in countries with lax or no relevant laws.

While the ECJ nuts this out, the proceedings concerning joint-tortfeasorship will continue. This too could make foreign websites liable under the law of the country where they are viewed, through joint liability with end users’ infringements.

The Web is big and it is sticky.

Much Ado About Kindergarten

I have been following for some time now - and with growing bemusement - a heated debate in Germany over the use of sheet music in nursery schools. The apple of discord is a letter from German collecting society GEMA, sent on behalf of another German collecting society (VG Musikedition, which represents composers, lyricists and publisher of sheet music) to some 36,000 German nursery schools. In said letter, GEMA reminds the nursery schools that under German copyright law you need a licence to copy sheet music and proposes the conclusion of a respective licence agreememt.

So far, so normal, one may think, but to most of the German press, this simple occurrence was a complete outrage. Emotions running high, most headlines read something along the lines of "Nursery Schools to Pay for Singing!" and "GEMA Rip-off in Kindergarten!" Inevitably, politicians jumped on the bandwagon: Sibylle Laurischk (of liberal democratic FDP) said that "singing in nursery schools is a basic part of education" and that, therefore, GEMA should exempt nursery schools from paying licences (http://www.bild.de/politik/2010/fuer-kinderlieder-15226246.bild.html). Heiko Maas (of social democratic SPD) echoed the "rip-off in kindergarten" view and said singing in nursery schools was an expression of an untroubled childhood (http://www.focus.de/finanzen/news/gema-kritik-an-kinderlieder-gebuehr_aid_585392.html). They and most other critics conveniently ignored that the letter did not actually concern the singing of any songs whatsoever. Unless nursery school children perform at a public event, they may sing whatever they like free of charge.

What the letter does concern is the copying of sheet music. German law only contains a very limited private copy exception for sheet music. According to s. 53 subs. 4 lit. a) German Copyright Act, one may only make a copy of a graphic recording of a musical work (1) by means of manual copying, (2) for the inclusion in a personal archive if and to the extent the reproduction is necessary for that purpose and a PERSONAL copy of the work is used to make the reproduction, or (3) if the copy is for personal use AND the work has been out of print for at least two years. All other uses require a licence from the right holder. GEMA (as instructed by VG Musikedition) now offers such licences to nursery schools for €56 per annum for up to 500 copies, € 112 for up to 1,000 copies and so on. So that's roughly €0.11 per copy (if you get a licence for up to 2,500 copies or more, the rate decreases slightly).

That is, if you actually make copies. I don't know about the esteemed readers of this blog, but when I was at nursery school, I learned songs by repeating from and singing along with my teachers. Most children aged three to six can arguably neither read lyrics nor musical notation, so the number of copies needed (if any) should in most cases be accordingly minor. Also, one may, of course, copy public domain works to one's heart's content. Concerning the arrangement of such public domain works, s. 3 sent. 2 German Copyright provides that "insignificant adaptations of a non-protected musical work shall not enjoy protection as independent works."

One good thing that has come out of this at times rather ludicrous debate is a collection of just such public domain songs suitable for children (which make me feel about four years old and can be downloaded here: http://data.musikpiraten-ev.de/public/kinder-wollen-singen.pdf). The initiators, Frankfurt-based "Musikpiraten e.V." have pledged to print some 50,000 copies of the booklet and donate them to nursery schools throughout Germany; the campaign is funded by charitable contributions from a range of companies and individuals who raised about €40,000 in less than two months (http://musik.klarmachen-zum-aendern.de/kinderlieder) - quite an achievement!

One maybe or maybe not so good thing is a draft bill for an amendment of the German Copyright Act submitted to the Bundestag by the parliamentary group of socialist party DIE LINKE ("The Left"). I shall mercilessly dissect it in a separate blog post tomorrow, so watch this space.

Bits and pieces

Further to John's post yesterday ("Liability in Italy -- Yahoo! held responsible", here), the text of the decision is now available, naturally in Italian, on Interlex.it, here.  Apparently the initial reports of this decision were somewhat overstated and the end of the world has not yet arrived for ISPs: the ruling is a 'procedimento cautelare', a sort of preliminary or summary judgment and not a full decision. No further reasoning will therefore be produced. The judge ordered Yahoo! to remove the links to infringing websites since the company had knowledge of the infringement because it received notices sent by the plaintiff (PFA Film srl) but had failed to activate the take-down procedure, became liable. Google Italia and Microsoft (who were also parties) were not found responsible since they did not administer their engines directly; they were held entitled to receive compensation of the costs of the judgment from PFA (thanks, Gaetano Dimita, for the link and the explanation).

This one seemed
quite nice, but
it has apparently
been taken ...

Having stuck with the portrait of Queen Anne (she of the Act of Anne) since April 2009, the 1709 Blog has decided to dispense with her services and get itself a real logo.  If you fancy your talents as a logo designer, please send your entry to Jeremy here with the subject line "1709 logo", and remember to let the blog have an irrevocable non-exclusive licence to use it on the blog and for its promotions. A prize, in the form of a copy of International Copyright by Paul Goldstein and Bernt Hugenholtz, will go to the best effort received by midnight on Sunday 3 April.

... or do they?

"Do bad things happen to works when they fall into the public domain?"  Further to Professor Paul J. Heald's thoroughly entertaining and instructive seminar last week (noted here), the 1709 Blog can report that Paul's PowerPoints are now available here.

Where it hurts most – right in the copyright bundle

Last week the New York Court of Appeals gave an intriguing answer to a jurisdictional question from the Second Circuit Court of Appeals. The case is Penguin v American Buddha (previous post here). The question and its answer are:

‘In copyright infringement cases involving the uploading of a copyrighted printed literary work onto the Internet, is the situs of injury for purposes of determining long-arm jurisdiction under N.Y. C.P.L.R. § 302 (a) (3) (ii) the location of the infringing action or the residence or location of the principal place of business of the copyright holder? In answer to this reformulated question and under the circumstances of this case, we conclude it is the location of the copyright holder.’

The New York Civil Practice Law and Rules give NY courts jurisdiction over non-domiciliaries who commit torts in NY and those who commit torts outside NY that cause injury to person or property in NY (the Big Apple’s equivalent of Shevill v Presse Alliance in the EU). Penguin alleges that copyright infringements were committed by the defendant in Arizona or Oregon, by uploading their books on servers there. The question is whether these out-of-state infringements caused injury in NY, where Penguin USA is based. The New York Court of Appeals explained their reasoning as follows:

‘Although it may make sense in traditional commercial tort cases to equate a plaintiff's injury with the place where its business is lost or threatened, it is illogical to extend that concept to online copyright infringement cases where the place of uploading is inconsequential and it is difficult, if not impossible, to correlate lost sales to a particular geographic area. In short, the out-of-state location of the infringing conduct carries less weight in the jurisdictional inquiry in circumstances alleging digital piracy and is therefore not dispositive. The second critical factor that tips the balance in favor of identifying New York as the situs of injury derives from the unique bundle of rights granted to copyright owners.… Based on the multifaceted nature of these rights, a New York copyright holder whose copyright is infringed suffers something more than the indirect financial loss we deemed inadequate in Fantis Foods. For instance, one of the harms arising from copyright infringement is the loss or diminishment of the incentive to publish or write…’

Basically what this appears to mean is since (a) the loss-of-sales injury caused by copyright infringements on the internet could be anywhere, it’s hopeless trying to pin that down but (b) the plaintiff’s rights themselves are injured (devalued) by the copyright infringement, it makes sense to say the injury happened where the plaintiff is located. This has a certain elegance, tidying up the ubiquitousness of internet copyright infringement. On the other hand, the idea that copyrights are located in the offices of a plaintiff’s business and are themselves injured is not only somewhat bizarre but may prove less than straightforward. What if the plaintiff is a licensee or there are several joint copyright owners in different locations?

Kazakhstan takes Steppes to prevent MS misappropriation

Kazakh Steppe (small extract)

News travels fast from the US and other hi-tech instant message jurisdictions, but it reaches the 1709 Blog in a more stately fashion when it comes from the Kazakh Steppe. Still, better late than never!  On 14 October of last year the Kazakhstan Ministry of Justice issued Order No. 279 regulating the protection of unpublished works, specifically the delivery, acceptance and storage of unpublished manuscripts.  A news item from Petosevic reports that

According to paragraph 3 of the Order, the Order entered into force ten calendar days after its first publication, which was on 12 January 2011 in the national newspaper Kazakhstanskaya Pravda. 

The new regulations were drafted in accordance with the Kazakhstan Law on Copyright and Related Rights and aim to prevent misuse and misappropriation of unpublished works. 

The Justice Ministry has appointed the Committee for Intellectual Property Rights as the authority responsible for protection of unpublished works.

How refreshing, when everyone else seems to be worrying out file-sharing and pursuing the Golden Grail of a copyright-proof business model for the digital age, that here at least is an issue that depends more on matters of principle than in trends of technology.  But what might have prompted this legislation at this particular time? A severe outbreak of manuscript-rustling?

Source: Kazakhstan PTO, via "Kazakhstan New Regulations on Protection of Unpublished Works", Petosevic

Liability in Italy - Yahoo! held responsible

We are grateful to one of our Italian friends for sharing with us the following information about a surprising decision of the Italian courts:

Yahoo! is responsible for all the copyright infringements in relation to unauthorized links: the caching providers’ responsibility PFA Films, an Italian production and distribution company, legitimate owner of all of the exploitation rights of the film “About Elly”, has summoned Yahoo! Italia S.r.l., the service provider, asking them to remove all the links and files unauthorized by PFA in connection to the film ”About Elly”. Surprisingly the Court of Rome agreed on the request and ordered Yahoo! to disable all the websites which allow the web surfer to watch the film or just some pieces of it in streaming, downloading or peer-to-peer. The judge basically says that only the official film website is lawful while all the others are illegitimate. 


The decision is really astonishing and probably it will be the beginning of a new trend. The judge indeed is very precise in the description of the facts and in focusing her attention in the interpretation of the law. She carefully referred to the legislation and to the interpretation of Italian courts in relation to the service providers’ responsibility. She rightly referred to art. 14, 15, 16 and 17 of decree 9/4/2003 n.70 and to the European Court of Justice (CGE C236/08 and C238/08) [the Google AdWords case] . She defined Yahoo! as a caching provider, which is responsible just for removing all the illegitimate links, only the ones that are brought out to its attention. So Yahoo! was responsible because it did not remove all the pirate links, yet PFA Films did not provide any lists of websites asking in such a general way for Yahoo! the clearing of every content reproducing all or just part of the film “About Elly”. The conclusion is a bit complicated to understand and takes the distance from the other previous decisions on he matter. Yahoo! must not previously verify the contents of the websites to which it links because it does not have any legal obligations to identify the contents previously, yet, just after a general warning notice by the owner of the exploitation rights of the audiovisual aids, it is responsible for removing all the pirate links or audiovisual files, because the official website of the film is the only one which is legitimated to show all or part of the film.

This takes the present UK discussions about a site-blocking remedy to a completely different place and is no doubt not the last word on this subject.

"Absurd" Limewire Damages Rejected

Following on Ben’s post earlier about the closure of Limewire and its effect on illegal filesharing, a note about the damages ruling recently issued by the judge in the Limewire case, judge Kimba Wood of Manhattan federal district court. Judge Wood rejected the plaintiff record companies’ demanded damages, as reported here. The plaintiffs’ damages request would have reached as high as $75 trillion under their theory of statutory recovery for copyright infringement. Judge Wood noted that this amount is "absurd" and is “more money than the entire music recording industry has made since Edison's invention of the phonograph in 1877.” Rather than assess damages for each instance of infringement, as requested by the plaintiffs, she limited damages to one damage award per infringed work.

While her reasoning makes practical sense in consideration of the fact that the astronomical damages sought by the plaintiffs could not possibly be satisfied by defendant payouts, it nonetheless creates a situation where plaintiffs winning infringement claims on a “smaller” level are able to maximize their damages with damage awards for each instance of infringement, but plaintiffs whose works have been infringed countless times by a defendant are limited in their recovery to prevent “absurd” damages. Is such a result fair or does it dilute the deterrent factor intended by the availability of statutory damages?

Google Book Settlement hits brick wall

On Tuesday, Circuit Judge Chin rejected the Google Book Settlement. Is this the end of the road for the Settlement or merely another inconclusive step in proceedings that have been ongoing for six years (see potted history at the end of this post if you would like a refresher)? The parties can appeal within 30 days but as the Scrivener’s Error blog explains, this is unlikely to succeed. If the judge’s opinion stands, how much room for manoeuvre does it give the parties to redraft the Settlement?

The opinion states:

‘While the digitisation of books and the creation of a universal digital library would benefit many, the ASA [Amended Settlement Agreement] would simply go too far. It would permit this class action – which was brought against defendant Google Inc. (“Google”) to challenge its scanning of books and displays of “snippets” for online searching – to implement a forward-looking business arrangement that would grant Google significant rights to exploit entire books, without permission of the copyright owners. Indeed, the ASA would give Google a significant advantage over competitors, rewarding it for engaging in wholesale copying of copyrighted works without permission, while releasing claims well beyond those presented in the case.…

‘In the end, I conclude that the ASA is not fair, adequate, and reasonable. As the United States and other objectors have noted, many of the concerns raised in the objections would be ameliorated if the ASA were converted from an “opt-out” settlement to an “opt-in” settlement…. I urge the parties to consider revising the ASA accordingly.’

On the face of it, this language does not amount to insisting on right owners ‘opting in’ but when the opinion is taken as a whole, this does seem to be almost mandatory. The most significant reasons for which the judge finds the Settlement unacceptable hinge on its ‘opt-out’ nature:

1. It is for Congress to create any mechanism for the exploitation of unclaimed books (including orphan works).
2. The original litigation was about display of snippets in Google Book Search – it is an unacceptable use of the class-action process that the Settlement releases claims in respect of the extensive further uses that are permitted under the Settlement.
3. ‘It is incongruous with the purpose of the copyright laws to place the onus on copyright owners to come forward to protect their rights when Google copied their works without first seeking their permission.’
4. The Settlement would give Google ‘a de facto monopoly over unclaimed works’.

There seem to be only two ways to fix these problems:

1. Make the Settlement ‘opt-in’, in which case it would not be a class-action settlement but the settlement of claims by a specific list of people. They, and only they, would agree to waive their claims against Google and to offer Google licences to exploit their works.
2. Keep the suit as a class action and draft a settlement that only covers past uses of works in Google Book Search. It would not grant future licences, not even ongoing licences for exploitation in Google Book Search.

Potted History of the Google Book Settlement

From 2004 certain libraries have allowed Google to scan their stocks. Google makes the public domain books available in full on Book Search. For in-copyright books a search returns ‘snippets’. Google say they do not need permission from the copyright owners for this: it is covered by fair use.

In 2005 Google was sued for copyright infringement by the Authors Guild (in a class action) and by five American publishers, subsequently deciding to settle, drafting the Google Book Settlement. In 2006 the La Martinière publishing group sued in France. In December 2009 the French court decided that Google Book Search did infringe French copyright. In April 2010 photographers, illustrators, graphic artists and photo libraries launched a suit.

The Settlement Agreement is far wider in scope than the original litigation over snippets. Google acquires US rights to sell ebooks, print-on-demand copies, online access; to include in institutional and consumer subscriptions; to make available free on library computers; and to return 20% of book in internet search. For in-print books, Google needs the express authorization of rightsholders but for out-of-print books this grant of rights is assumed and rightsholders must opt out.

Since it is the settlement to a class action, the Settlement can only be approved by the court after a hearing and on finding that it is fair, reasonable and adequate. The court gave the original Settlement preliminary approval in 2008, fixing the hearing for 2009. An avalanche of objections to the settlement were submitted to the court and the US government rejected it. The court therefore allowed the parties to submit an amended settlement.

The most significant change in the Amended Settlement was that it basically only included books published in the US, UK, Canada and Australia, instead of the whole world. The Settlement had received strong opposition from Continental Europe, especially France and Germany, but barely a murmur from the UK. It is obvious that the revised geographical scope of the Amended Settlement was designed to cut out opposition from the main critics. The US government again rejected the Amended Settlement but expressed willingness to continue to work with the parties. At the hearing on 18 February 2010, the judge said he was not going to rule. His judgment was reserved – until 22 March 2011.

During these protracted proceedings, Google has continued to scan books at an ever-increasing speed. In October 2008, when the Settlement was first drafted, Google had scanned 7 million books. A year later 10 million books had been scanned. Now the figure has passed 15 million.