The team is joined by GuestKats Mirko Brüß, Rosie Burbidge, Nedim Malovic, Frantzeska Papadopolou, Mathilde Pavis, and Eibhlin Vardy
InternKats: Rose Hughes, Ieva Giedrimaite, and Cecilia Sbrolli
SpecialKats: Verónica Rodríguez Arguijo (TechieKat), Hayleigh Bosher (Book Review Editor), and Tian Lu (Asia Correspondent).

Thursday, 22 March 2018

National and EU text and data mining exceptions: room for coexistence?

If only all mining activities
were these cheerful!
A few days ago this blog reported that Irish Government has recently proposed the introduction into Irish law (new section 53A of the Copyright Act) of a copyright exception for text and data mining (TDM).

From an EU standpoint, the most interesting aspects of the Irish proposal are probably the following:
  • First, its timing: while it is true that a TDM exception was first proposed in the 2013 Modernising Copyright Report, an actual initiative of that Member State's Government comes at a time when discussion is well underway to introduce a mandatory TDM exception at the EU level [Article 3 of the Proposal for a Directive on copyright in the Digital Single Market - on the EU proposal see further here and here]
  • Secondly, the different scope of the Irish exception [which would be derived from Article 5(3)(a) of the InfoSoc Directive] and the EU exception, in the sense that the former – while only allowing non-commercial TDM – would not limit the catalogue of beneficiaries. The EU exception – at least in the version proposed by the EU Commission - would instead be available to ‘research organisations’.
Following the publication of the IPKat post, I have engaged in some discussions on whether and to what extent national TDM exceptions – having a different scope from the EU one – would survive after the introduction of an EU exception. Another question would be whether EU Member States could continue to rely on Article 5(3)(a) of the InfoSoc Directive to introduce their own TDM exceptions following the introduction of an EU exception.

These questions do not have an easy answer, especially if one looks at the text of the directive as originally proposed by the EU Commission.

Dreaming of freedom ...
The Commission’s version

Arguably, the text of the original Commission’s proposal does not leave room for national TDM exceptions.

If in fact one looks at Recital 5, the goal seems to be one of maximum harmonization, in the sense that other directives – including the InfoSoc Directive and its TDM exception under the umbrella of the research exception – would only apply to “uses not covered by the exceptions or the limitation provided for in this Directive [ie the Directive on copyright in the Digital Single Market]”. 

This might mean that, as the new directive covers TDM, there will be no room for national transpositions of Article 5(3)(a) of the InfoSoc Directive to allow TDM that bypass the Directive on copyright in the Digital Single Market.

This conclusion appears reinforced if one also considers Recital 14, which highlights the shortcomings of TDM exceptions rooted within the InfoSoc Directive alone.

The Council’s latest version

However, things may be different if the version of the EU TDM exception eventually adopted was the one most recently proposed by the Council.

In fact, in this version (which would appear the last instance in which the Council has touched upon TDM), what is interesting is the wording of Recital 5 and Article 17.

First, Recital 5 states that “[t]he exceptions and limitations existing in Union law should continue to apply, including to text and data mining, education and preservation activities, as long as they do not limit the scope of the mandatory exceptions laid down in this Directive and on condition that their application does not adversely affect nor circumvent the mandatory rules set out in this Directive”. 

This would suggest that national TDM exceptions could survive after the adoption of this new directive, as long as certain conditions are satisfied. In a way, this solution appears to have points of resemblance with the grandfather clause in Article 5(3)(o) of the InfoSoc Directive.

... and freedom
Article 17(2)(b) requires that the InfoSoc Directive is amended, to the effect that exceptions adopted pursuant to Article 5(3)(a) will be without prejudice to the exceptions and the limitation provided for in the new directive … including TDM. 

This might suggest that, even after the introduction of an EU TDM exception, Member States could retain the freedom to introduce their own national (non-commercial) TDM exceptions, without any particular limitations as regards the types of beneficiaries.

While an approach of the kind suggested by the Council would have the merit of retaining some flexibility and national freedom as regards possible approaches to and regulation of TDM activities, it could also impair the goals of harmonization, including underlying internal market-building goals. In any case, an EU action of this kind would achieve minimum harmonization in the area of TDM (especially if the resulting EU exception was only aimed at research organisations), while substantial disparities might remain in place (or subsequently arise) at the EU level.

Next steps

The Bulgarian presidency of the Council is expected to continue its work and find compromise solutions that meet the approval of Member States. 

In parallel, the European Parliament is also working on the text of the directive, and the JURI Committee Rapporteur – MEP Voss – is expected to issue his Report sometime soon, the vote in that Committee being now scheduled for late April.

Courts continue to struggle to find appropriate forms of injunctive relief when social media is involved: Frank Industries versus Nike

Social media offer new opportunities for trademark owners to contact and engage consumers.  At the same time, alleged unauthorised uses of trademark by third parties in social media threads have become more and more frequent. In an attempt to control and limit the damage, rightsholders apply for injunctive relief, presenting the courts with the difficult task of balancing between the protection of a registered trademark and the irreparable damage that might result from erasing social media threads in connection with the alleged infringer.  The case at hand, Frank Industries v. Nike Retail [2018] EWCA Civ 497 , in addressing the alleged unauthorised use of the registered trademark LNDR (UK and EU mark) owned by Frank Industries (FI), by Nike UK, provides an excellent example of this tension. 

Both companies design and sell sportswear. Nike launched an advertising campaign to promote its brand to Londoners, under the slogan “Nothing beats a Londoner”. During the campaign, Nike used the marks LDNR and LNDR as forms of abbreviation for the slogan. The campaign was launched in the beginning of January 2018, and already on 26 January 2018, FI's solicitors wrote to Nike, complaining that the sign was infringing the registered mark.  Nike carried on using the sign, and the following week it proceeded to launching a video using the signs on YouTube. 

FI began proceedings against Nike, alleging both trade mark infringement and passing off seeking also both an injunction restraining Nike from infringing the marks and from passing off and an order for delivery up or destruction on oath of infringing material.

Preliminary injunctions ruling

On 22 February 2018, FI applied for an interim injunction, which  was granted on  2 March 2018 by the High Court of Justice ([2018] EWHC 424 (IPEC). Paragraph 1 of the order was a restraining order while paragraph 2 was a mandatory order requiring Nike, by 4 p.m. on 16 March 2018, to take all reasonable steps to delete the signs LDNR, LNDR, LDNER and LNDER from social media accounts within its reasonable control, including Twitter, Facebook, Instagram and YouTube.  The scope of the interim injunction is for a period of four months (taking into consideration that the trial date is set for the 12th of July 2018). The ruling of the High Court was appealed and the Court of Appeal ruling came on the 13March 2018.

Lord Justice Lewison considered the steps followed by the High Court in ruling on the preliminary injunction (see paragraph 10 of the High Court judgement):

 "Following the sequence set out by Lord Hoffmann, I must consider the following matters in deciding whether to grant an interim injunction.  One, whether the damages will be an adequate remedy for the claimant; if so, no injunction should be granted.  This must be on the assumption that the defendant is good for the damages: See American Cyanamid at page 408. Two, whether there is a serious question to be tried; if not, no injunction should be granted.  Three, if the answers to one and two are no and yes, respectively, whether the cross‑undertaking in damages would provide the defendant with an adequate remedy if an injunction were granted; if yes, the injunction should be granted.  Four, if there is a risk of irreparable harm both ways, the court must go on to assess the balance of irreparable harm in order to determine whether granting or withholding the injunction is more likely to cause the greater irremediable prejudice overall.  Five, among the non‑exhaustive list of matters which the court may take into account are (a) the nature of the prejudice to the claimant if no injunction is granted and the nature of the prejudice to the defendant if an injunction is granted; (b) the likelihood of the prejudice in each case occurring; (c) the extent to which the prejudice may be compensated by an order of damages, or by the cross‑undertaking; (d) the likelihood of either party being able to satisfy such an award; and (e) the court's opinion of the relative strength of the parties' cases."

According to Lord Justice Lewison, the High Court application of the above-mentioned criteria was correct. There is a serious danger that the public would perceive FI’s own goods as being in some way part of Nike’s business. The issue to be tried was undoubtedly serious, and thus the  remaining core question was whether Nike would suffer irreparable harm by the grant of the injunction. While the reasoning behind the prohibitory injunction is, according to the Court of Appeal, solid, the High Court  sparse in its words when it comes to the mandatory injunctions. It simply provided that Nike is not only supposed to refrain from future activities but also to take affirmative steps to remove Instragram posts, tweets and a Youtube video.

According to the Court of Appeal, mandatory actions in general are more likely to cause irremediable prejudice than where a defendant is merely prevented from taking or continuing with some course of action (see Films Rover International Limited v Cannon Film Sales Limited [1987] 1 WLR 670. The High Court judge dealt with the mandatory part of the order in one short passage, at paragraph [28] of the judgment.  He said: 

"The effect of an injunction would be to require Nike to remove its video in its present form from YouTube and other media and to remove LDNR references from its website.  So far as the YouTube video is concerned, LDNR is only used, or only significantly used, right at the end of the video.  Mr. Campbell told me that no objection would be taken to the remainder of the video being used. Therefore all that Nike would have to do would be to remove that small last section; otherwise, the video can remain on YouTube and other social media.  So far as Nike's website is concerned, it would not, it seems to me, be very difficult to remove the sign."

The ruling of the Court of Appeal

The Court of Appeal had to consider the  delicate balance between the interest in providing adequate and efficient injunctions until trial, and the importance of not causing irreparable damage to the social media presence of the alleged infringer. The Court of Appeal concluded that the High Court ruling disregarded the fact that amending a YouTube video entails its removal and reposting (with a different URL).  The effect would be that Nike would lose all posted comments as well as all video shares and likes. Similarly, deleting Instagram posts would lead to the removal of all related comments and posts. 

Twitter posts particularly concerned the Court of Appeal. Deleting existing posts was considered by the Court of Appeal to be of an irreversible and far-reaching nature because it would deprive Nike of the benefit of continuing conversations between young Londoners. 
LNDR an abbreviation for London, free for all or trademark use?

Twitter posts particularly concerned the Court of Appeal. Deleting existing posts was considered by the Court of Appeal to be of an irreversible and far-reaching nature because it would deprive Nike of the benefit of continuing conversations between young Londoners. The Court of Appeal  reversed thus the High Court ruling   regarding Twitter posts. Furthermore, the Court of Appeal ruled that with regards to Instagram, Nike would be able to archive them rather than simply deleting them, while the YouTube video need not be removed provided the offending signs are blurred.

This case reminds us of the importance of social media, the role they play and the power that they have in the communication between trademark holders and the public. They are much more than a mere communication media, being rather the platform upon which public discussion takes place and, as such,  worth protecting (such as the tweets by young Londoners in connection with Nike). The Court of Appeal, contrary to the High Court, seems to take into consideration the potential value of social media (by providing for the archiving posts in the case of Instagram and blurring offending signs in YouTube videos) as well as  undesirable  technical  consequences (such as deleting a tweet deletes all subsequent comments and discussions on the  tweet in question) in order to find a balance between the interests of the rightsholder and the potential damages inflicted on the alleged infringer.

Wednesday, 21 March 2018

Controversy looming for this year's pre-EQE

The pre-examination European Qualifying Examination ("pre-EQE") was introduced in 2012 as an entrance exam to the main EQE for European Patent Attorneys. Only after passing the pre-EQE is a European patent attorney trainee allowed to enroll for the Main Exam. 

Controversy over the pre-EQE is not new. In previous years, candidates have successfully appealed to the disciplinary board of appeal (DBA) against their result on the grounds that the answers to some of the questions were ambiguous (D3/14)

This Kat has been alerted to the fact that, this year, the questions for the pre-EQE were particularly ambiguous, setting the stage for the greatest pre-EQE controversy so far. 

UPDATE: The pre-EQE results have just been released, see here.  We still await release of the Examiner's report. 

2018 pre-EQE

The pre-EQE

The pre-EQE is a multiple choice exam, with 20 questions, each question having 4 statements for which candidates must indicate true or false. A complicated marking scheme dictates that for each question: 4/4 = 5 marks, 3/4 = 3 marks, 2/5 = 1 mark, and 1/5 = 0 marks. The complicated mark scheme highly rewards getting all of the statements within a question right, or equivalently, heavily penalises one mistake in a question. The pass mark is 70/100. 

The paper is divided into two sections, legal questions and claim analysis. The claim analysis section deals with issues of clarity, added matter, novelty and inventiveness with regards to a series of claim sets, amendments and prior art documents.  

It may seem odd to those used to EP patent application prosecution, that claim analysis may be reduced to mere true or false statements. Deciding on the closest prior art and inventiveness without the opportunity to argue your point can feel, at times, unfair. At the very least, it is hoped that the issues presented in the exam should be fairly clear-cut and unambiguous. 

To complicate matters, the pre-EQE may be taken in English (EN), French (FR: "Examen préliminaire") or German (DE: "Vorprüfung"), for which there are corresponding versions of the paper. It is intended that the answers to the EN, FR and DE papers are identical. 

Every year, DeltaPatents provides their model solution to the pre-EQE (EN version) before release of the official mark scheme. In recent years, the German Plattform für den Gewerblichen Rechtsschutz (GG-IP) has also been providing model solutions (DE version). It has become customary for candidates to check their solutions with these model answers before the release of the official mark scheme and results. In 2017, the DeltaPatents solution was identical to the official solution.


Given the subject-matter and format of the exam, it is perhaps not surprising that the pre-EQE has already been beset by controversy. In 2015, following successful appeal by a candidate to the disciplinary board of appeal (DBA), the EPO published an Addendum to the 2015 pre-EQE, in which it was decided to a award marks for either True or False for 2 statements in Question 15 and 17 respectively. After some initial uncertainty, the marks of candidates who did not file an appeal, but who would have passed in light of the changes, were upgraded (IPKat here and here).

This year

The 2018 pre-EQE (EN, FR and DE) took place at the end of last month. This Kat has been alerted by colleagues to a worrying disparity between the solutions provided by GG-IP and DeltaPatents this yearParticularly, the GG-IP and DeltaPatent model answers differ on 5 points, corresponding to an alarming 10 marks (out of 100), (for comparison, last year's GG-IP solution differed from the DeltaPatents solution on 2 statements, corresponding to 4 marks out of 100). The complicated mark scheme of the pre-EQE causes ambiguity in a few statements to have a disproportionately large impact on a candidate's marks. The official solution has not yet been released. 

The disagreement was in regard to the answers to questions 11.2, 12.4, 14.3, 16.1 and 18.1. 
Pre-EQE 2018: Water filter lid

All of these questions are also highlighted as problematic in the DeltaPatent pre-EQE blog, and from a quick analysis of the paper, this Kat agrees that the answers to these questions are not clear cut (at least in the EN version). Q12.4, for example, relates to Article 84 EPC (clarity) and asks whether a feature of a claim simply claims a solution to a problem and not the technical features that provide this solution:

"...a counter configured to automatically increment in response to water being added to the jug"

DeltaPatents thinks no, Article 84 EPC is not contravened, whilst GG-IP thinks that it is. This Kat notes that somebody involved in the prosecution of European patent applications, might expect the EPO to reject such a claim, but that the Examiner may be persuaded by argument. 

Q16.1 relates to claims directed to a plurality of interrelated products. GG-IP concluded that 2 independent claims (a 1st claim directed to a filter lid for a water jug and a further claim directed to a jug with a filter lid), are interrelated products, whilst DeltaPatent believes that they are not. 

Worryingly, still further issues with the claims have been raised in the comments on the DeltaPatent blog (that do not even correspond to the differences between Delta and GG-IP answers). A particularly problematic question, according to DeltaPatents itself, is Q 12.2 which concerns the correct two-part form for a claim in view of a stipulated prior art document. 

The DeltaPatent and GG-IP bloggers are experienced European Patent Attorneys and tutors. The disparity between their solutions clearly demonstrates that the claim analysis section of this year's pre-EQE is far from unambiguous. 

Kat confused by claim analysis
Whilst the solutions to the EN, FR and DE versions of the paper should be identical, translation issues can lead to different interpretations. Indeed, according to D 1/94 a translation error may be considered to contravene Article 11(3) of the Regulation on the European qualifying examination for professional representatives (REE) "since this provision assumes that the translation from the language selected by the candidate into one of the official languages of the EPO is totally correct". However, such a magnitude of issues with this year's paper suggests this is not just a translation problem.

This Kat has been unable to find a model answer for the FR version of the paper, but feels this would be a useful comparison. 

As a gateway to the main EQE, passing the pre-EQE is of considerable importance to a trainee European patent attorney. A fail means a candidate must wait about another 2 years before they can take the Main Exam. The degree of ambiguity in this year's questions will thus be causing significant anxiety. 

The official pre-EQE marks scheme and results will be published in the coming months. Stay tuned to IPKat for further news. 

Tuesday, 20 March 2018

The 5 Pointz case: a response (on the risk of cultural gatekeeping in copyright)

The piece of (street) artwork at stake
in the '5 Pointz' case
The recent decision in the 5 Pointz case [here] has brought back to centre stage one of the quirks of US copyright law: the right to object to any intentional or grossly negligent destruction of works of ‘recognized stature’ (US Code, Title 17, §106). As explained by Katfriend Mira Sundara Rajan in a recent commentary of the case [here], this right was introduced by the 1990 Visual Artists Rights Act (VARA) under the umbrella doctrine of moral rights (although this right goes beyond what the moral right doctrine traditionally entails).

From one vantage, the decision is to be welcome, inasmuch as  it protects a notable piece of street art [per the court  and mainstream media, here and here], thereby  enabling the preservation  of more contemporary forms of our cultural heritage. However, this Kat has some reservations regarding the ability of the IP law framework to embrace such a role. In particular, who determines what is ‘recognized stature’ and under criteria is it determined? The answer: mainly as the court decides (read: judges), for the simple reason that Congress offered little in the way of definition or guidance on this point [see Cheffins at 600]. This places US judges in the difficult position of acting as cultural gatekeepers, given that they have the final say as to what is worthy  of such protection.
Judges themselves have noted  the challenge in being asked to rule on the protection of works of recognized stature [see Martin v City of Indiana (1999) ; Cheffins v Stewart (2016)]. This is especially so in that  they are called upon to render judgment on an issue in which they are not expected to have any  necessary expertise, namely, when is a work of "recognized stature”.
Some may argue that a concern about judges serving as cultural gatekeepers is allayed by the fact that the context is unique to US law and it is an unusual form of moral right. Moreover, since few decisions are actually rendered under this provision, this risk of cultural gatekeeping is well-contained. However,  in this Kat’s view, this it is not the only occasion in which judges are in the position of playing the part of cultural gatekeepers. After all, copyright law applies selectively to prescribed categories, and only these categories, of original creative works in a fixed form – conditions which are themselves culturally-bounded [here and here].
Furthermore, once copyright subsists in a work, it provides a platform to control its dissemination, thereby fundamentally impacting on the nature of our cultural landscape. Right holders, artists and their heirs can even pursue claims bordering on strategies of ‘heritage preservation’ or ‘heritage safeguarding’, either through the moral right doctrine of integrity, or the economic right to control some most derivative works, and of course, the US right to object to the destruction of works of recognized stature. Cultural gatekeeping thus can be seen as underpinning almost every aspect of copyright long before the right to object to the destruction of works of ‘recognized stature’ was introduced in the US.
Thus, the larger question is how to make sure no (or limited) cultural gatekeeping takes place there? First, we do so by reiterating  that copyright law does not deal with cultural quality or aesthetic considerations, something both commentators and  judges in various jurisdictions (including in the US) have often emphasised [see Gracen at 304; Schott Musik (1997) 145 ALR 483, 486, Hensher v Restawile (1974) 2 All ER 420, 423]. In many ways, eschewing any consideration of cultural or aesthetic quality is one of the main safeguards within copyright law to prevent cultural gatekeeping. Indeed, judges have been vigilant not to behave as art critics [see Bleinstein at 251]. When they have, they have been corrected thanks to the appeal mechanism, i.e. corrective mechanisms built into our legal systems.

A notable example was  the Victor Hugo decision (2003) in which the Court of Appeal of Paris had extended the moral  right of integrity enjoyed  by  Hugo’s heirs  so as to ban the publication of sequels of Les Miserables (focusing on the main character Jean Valjean).
The Paris Court of Appeal stated:
"… [O]utlawing sequels of Les Miserables does not, as the parties contend, breach the principle of free creation since, in the facts presented before us, this work, being a  pillar of world literature. [Les Miserables] […] is not a mere novel for it presents a philosophical and political approach”
“Considering that, it follows that no sequel shall ever be given to a work such as Les Miserables, forever complete, and that the company Plon has, by editing and publishing Cosette ou Le Temps Des Illusions and Marius ou le Fugitif, by presenting the works as sequels to Les Miserables, violated Victor Hugo’s moral right as vested in the latter literary work” [Kats’ translation]
This decision was subsequently overturned by the Court of Cassation [here and here], the highest court in France for matters of civil law, which stressed that the cultural significance of the work could not be taken into consideration in construing the law or in balancing the fundamental rights at stake.  
But are these safeguards enough? The appeal process suffers from a major drawback: it depends upon the claimants’ financial ability to pursue appeal proceedings. Even more, what about more subtle cultural references made in connection with determining the subsistence or infringement of copyright, another slippery slope of cultural gatekeeping? 

Thus,  in Hadley v Kemp (1999), Park J draw a (bold) comparison between the creative process followed by members of the Spandau Ballet and that of  Beethoven to, it seems, reach his  conclusion on the difficult distinction between authorship, joint-authorship and performership in copyright law ([1999] EMLR 589, 645-6).

The Spandau Ballet

The decision reads:
After all, when Mr Kemp [Spandau Ballet’s lead composer and musician] devised the song he devised it for performance, not by himself as a solo artist, but by Mr Hadley [Spandau Ballet’s lead singer] and the whole band. A composer can ‘hear’ the sound of his composition in his mind before he ever hears it played. Beethoven could hear his music in this sense even when he was deaf. When Mr Kemp was devising his songs the sound which he had in his musical consciousness must surely have been the sound they would have when performed by Spandau Ballet, not the sound they would have when sung by Mr Kemp alone to the accompaniment just of his own guitar.’

Beethoven and his cat unimpressed
by the comparison
Why use a canon of classical music to inform a decision on copyright about pop music marks?  The analogy drawn here by the Court between Beethoven and the Spaudau Ballet (or rather between their respective practices) begs the question of how much this comparison weighed on the outcome of the case. Would the court have sided with Kemp had the decision been informed by another cultural reference, say Björk, Aretha Franklin, Daft Punk or…Kim Kardashian?
There may be a risk of indirect cultural gatekeeping in shepherding a select few fountainheads of our culture as relevant illustrations for the ‘hows’ and ‘whys’ of creativity for copyright purposes. This question is particularly important when it affects the distribution of rights between claimants or the eligibility protection of creative expressions for copyright protection.  In this regard, Hadley v Kemp is not an isolated instance (see Rockford Map Publishers at 148-9; Miller v Civil City of South Bend at 1093-5; Garcia v Google at 742-3).
What could be the safeguard against cultural gatekeeping? The answer is discussing, disputing and engaging those who favour a select few canons of so-called ‘high-culture’ in copyright law? To be clear— the result of this process is not predetermined. It is only through critical inquiry that we can reach reasonable positions on whether cultural gatekeeping and the biases that it entails, are welcome parts of the copyright law, or   is it an inevitable deviance of copyright law-making processes?

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':