Thursday, 21 February 2019

Facebook found liable for hosting links to unlicensed content

Kilari and Valentina Ponzone
For the first time under Italian law a platform (Facebook) has been found liable for hosting third-party links to unlicensed content. 

The decision was issued by the Rome Court of First Instance a few days ago: it is Tribunale di Roma, sentenza 3512/2019.

The judgment is both interesting and important, also considering the YouTube referral (C-682/18) currently pending before the Court of Justice of the European Union (CJEU) [Katpost here]

Let's see more in detail what happened and how the Rome court reasoned.


Broadcaster Reti Televisive Italiane SpA (RTI) and singer Valentina Ponzone sued Facebook over the creation, on the latter, of a profile that hosted unlicensed videos and defamatory comments relating to Ms Ponzone's performance of the intro to the Italian version of Japanese animated TV series Kilari. The profile also contained a photograph of Ms Ponzone dressed up as character Kilari, and links to videos  hosted on YouTube and consisting of extracts from the animated TV series.

Facebook received 5 requests to deactivate the profile at issue, as well as removing all images, videos, and links. The relevant profile was only removed in 2012, ie 2 years after the first request had been sent.

RTI and Ms Ponzone submitted that Facebook would be liable for defamation and violation of image rights, as well as infringement of RTI's exclusive rights over Kilari and trade mark 'Italia 1'. They also claimed damages for EUR 500,000 (EUR 250,000 for Ms Ponzone and EUR 500,000 for RTI).

Facebook responded that, first, the Italian court would have no jurisdiction and, second, that no liability could subsist, it being eligible for the safe harbour protection under Article 16 of the Italian Legislative Decree 70/2003 (corresponding to Article 14 of the E-commerce Directive) and having no monitoring obligation as per Article 17 of the Decree/Article 15 of the Directive.

The decision

The Rome court dismissed both arguments raised by Facebook.


With regard to jurisdiction, the judge found that Article 5(3) of the Brussels Convention would apply. The place where the harmful event occurred must be intended not as the place where the infringing content was uploaded, but rather as the place where the damage is felt, that is where the victim resides or exercises its economic activity. In this case, this place would be Italy. This conclusion would be in line with CJEU case law on the Brussels I Regulation recast, in particular the decision in Hejduk [here].

Thus, the relevant jurisdiction clause in Facebook's Terms of Service would be inapplicable to the case at issue, as it relates to contractual disputes between the platform and its users, not actions brought in tort.


With regard to the content of the comments, the court concluded that they had defamatory character. 

Of particular interest is the conclusion reached with regard to the publication of links. By referring to relevant CJEU case law - in particular the recent decisions in Filmspeler [hereand Renckhoff [here- the Rome court held that [the translation from Italian is mine]:
The publication of RTI's audiovisual content through Facebook is an act of communication to a new public in that it is a public other than the one authorized by the claimant. Indeed, the links published through the Facebook page led not to content published by RTI itself through its own platform, but rather content published through a third-party site (YouTube) not authorized by RTI to making available the audiovisual content at issue. It follows that, lacking a specific authorization by RTI, the making available to the public (through a third-party portal) of the intro to animated series 'Kilari' must be considered unlawful.
The above appears to suggest that to display content on YouTube the relevant rightholder's permission would be needed, but the court failed to elaborate on this (important) point.

What the court discussed a bit more at length is (1) how the infringing content is to be notified and (2) the liability of Facebook.

The Rome court first recalled that the safe harbour for hosting providers only applies to providers that are passive. If a provider acquires knowledge - ex post - of the infringing activity, it has a duty to act expeditiously to remove or disable access to the relevant (notified) content.

According to the Rome court [this is not the first time: see, eg, here] to notify a provider of an infringing activity it is not necessary to submit the relevant URL for each and every infringement. The first request that RTI submitted contained a URL to the Facebook page where it was possible to subscribe to a group that would allow one to read the relevant comments and view, by clicking on the links, the videos. In any event, [again, the translation is mine]:
the indication of the URL is technical information which does not coincide with the individual infringing contents available on the digital platform, but only represents the 'location' where such contents are available and, as such, is not an indispensable element for their localization
According to the Rome court, Facebook did not act in accordance with the level of diligence required for a hosting provider:
The operators of the Facebook portal failed to adopt all measures that could be reasonably expected in the case at issue to prevent the unlawful making available of content and, as a result, they failed to act in accordance with the diligence that might be reasonably expected from the hosting provider
As such, the court found Facebook liable - by omission - for violating the personal rights of the claimants, Ms Ponzone's image rights, and infringing RTI's exclusive rights over the audiovisual content. No trade mark infringement was found as Facebook had not 'misappropriated' the registered sign. The court also awarded damages for little over EUR 30,000.

AnimeKats: Sailor Moon's Luna

The decision is interesting for a number of reasons, but in particular in relation to the issue of knowledge of a hosting provider of third-party infringing activities.

One of the questions referred in YouTube is indeed the following [the translation from German is mine]:
Does actual knowledge of the unlawful activity or information as referred to in that provision [Article 14(1) of the E-commerce Directive] and the awareness of the facts or circumstances from which the illegal activity or information becomes evident have to be in relation to specific unlawful activities or information?
This question seeks clarification on the notion of ‘knowledge’ (and ‘awareness’) within Article 14(1) and also, as it appears, the degree of specificity of a takedown request. The CJEU has not yet had a chance to provide specific guidance on this point. However, from the decision in L’Oréal, there appears to be a duty of specificity on the side of the rightholders. In that case the CJEU suggested that notifications of allegedly illegal activities or information that “turn out to be insufficiently precise or inadequately substantiated” could not preclude the availability of the safe harbour to the provider.

In addition, one might wonder whether a duty of the hosting provider to remove or disable access to infringing content even absent knowledge of specific unlawful activities or information might be contrary to Article 15 of the E-commerce Directive. In this sense, in its twin judgment in Scarlet and Netlog,  the CJEU found that SABAM’s request to the providers at issue in those cases (respectively, an access and a hosting provider) to prevent the making available of works in its repertoire would be tantamount to imposing a general monitoring obligation.

The reasoning of the Rome court does not instead appear too helpful with regard to what is probably the KEY question in YouTube, that is whether the operators of a platform of this kind could be regarded as directly liable for the content uploaded by users ... à la Pirate Bay [Katposts here]

Indeed, the Italian court appeared to focus more on the dichotomy safe harbour availability / unavailability, rather than a direct responsibility (and, therefore, liability) of Facebook operators for - inter alia - linking to infringing content. Nonetheless, the conclusion of the Rome court was that Facebook operators would be liable - inter alia - for infringing RTI's rights under Article 79 of the Italian Copyright Act.

In all this, it will be crucial to see what the CJEU decides in YouTube, although the question of liability might become less pressing. When the EU court decides the referral, at the policy and legislative level we may indeed have an answer already - in the affirmative (certain platform operators are directly responsible for the making available of user-uploaded content). The new Directive on copyright in the Digital Single Market might be indeed at the national transposition stage by that time. 

[Originally published on The IPKat on 21 February 2019]

Saturday, 9 February 2019

'CHIARA FERRAGNI' not confusingly similar to 'CHIARA', rules EU General Court

Those who follow all things influencers will be pleased to learn that yesterday the General Court (GC) issued a ruling (T-647/17, only available in Italian and French) in which it found that the application to register the figurative sign ‘CHIARA FERRAGNI’ as an EUTM could proceed.

What happened

In 2015 Serendipity Srl, a company of which fashion blogger, IT girl and influencer Chiara Ferragni is a member, filed an application to register this sign as an EUTM in classes 18 and 25:

The application was opposed in accordance with Article 8(1)(b) EUTMR by a Dutch company that holds earlier word Benelux trade mark 'CHIARA' registered in class 25.

The EUIPO Opposition Division decided in favour of the opponent in respect of certain goods in class 18 and for all goods claimed in class 25, on grounds that likelihood of confusion would subsist between the 'CHIARA' and 'CHIARA FERRAGNI' signs.

The applicants appealed the decision to the EUIPO Fourth Board of Appeal, but without success. The Board found in fact that the signs were visually similar to an average extent and phonetically similar above average. The conceptual similarity was ‘neutral’. The Board also found that ‘CHIARA’ possessed a ‘normal’ intrinsic distinctiveness.

A further appeal to the GC naturally followed. Here's how the court reasoned.

Comparison between the signs

Having noted that no issues subsisted as regards the identification of the reference public and the fact that the goods covered by the marks were identical, the GC recalled the general approach that needs to be adopted when assessing similarities between two signs.


With specific regard to the trade mark applied for, the GC held that the figurative element of ‘CHIARA FERRAGNI’ represents an eye that has a peculiar stylization. As such consumers would perceive it as an elaborate and original element, which may be easily memorized. Hence, it cannot be regarded as merely an ornamental element. In addition, as it does not have any connection with the goods claimed, the figurative element also possesses an intrinsic distinctive character. Furthermore, it being placed above the word element and having a much bigger size than that, it has at least a position of relevance equal to that of the word element.

In light of all this, it would be incorrect to conduct the assessment of similarity based only on the word elements of the two signs.

Visual similarity: low

Having recalled that word and figurative signs may be compared for a finding of similarity and that the fact that an earlier trade mark is included in its entirety in a later mark hints towards a finding of similarity, the GC stated that the issue was to determine whether any such similarity would be average (as the Board held) or, rather, weak (as the applicants claimed).

The GC found that the figurative element in the ‘CHIARA FERRAGNI’ sign is as important as the word element, and contributes to the overall visual impression thereof. In addition, the element ‘FERRAGNI’ is visually more important than ‘CHIARA’.

It follows that the earlier trade mark ‘CHIARA’ and the sign ‘CHIARA FERRAGNI’ should have been considered similar to a low extent.

Chiara Ferragni and husband Fedez -
is a recently registered EUTM (017935134)
At this point the GC warned against an ‘automatic’ application of certain criteria established in case law. So, for instance, whilst it is true that in principle the initial part of a sign is more relevant than the final part, this is not always the case. What matters is the overall impression conveyed by the signs at issue, since the average consumer normally perceives a sign as a whole and does not examine the individual elements thereof. 

In this case, the fact that the element ‘CHIARA’ is placed before ‘FERRAGNI’ cannot exclude that the overall impression conveyed by this sign and the earlier trade mark is similar to a limited extent.

Also the idea that the name ‘Chiara’ would sound somewhat “exotic” to the reference public (Benelux) does not make up for the fact that the visual similarity between the signs at issue is weak.

Aural similarity: average or low

Turning to the assessment of the aural similarity, the GC noted that:
  1. ‘FERRAGNI’, given its length, is more relevant – aurally – than the element ‘CHIARA’. 
  2. It cannot be presumed that the element ‘FERRAGNI’ would not be pronounced. 
  3. One should avoid – once again – any automatism and think that just because ‘CHIARA’ comes before ‘FERRAGNI’, it is more important than this.
As such, also the aural similarity between the signs ‘CHIARA’ and ‘CHIARA FERRAGNI’ should be regarded as average if not even low.

Conceptual similarity: non-existent

Turning to the final type of similarity assessment, the applicant submitted that ‘CHIARA FERRAGNI’ enjoys a repute that ‘CHIARA’ does not possess, in that it is the name and surname of “fashion blogger Chiara Ferragni, known all around the world as an influential person in the fashion field and already well-known at the time of filing the application to register” [para 63; the translation from Italian is mine].

In support of this, the applicant provided an extract from Wikipedia, a copy of Chiara Ferragni’s Instagram profile and several extracts of publications in the fashion sector in different countries. 

According to the opponent, this evidence would be insufficient to prove knowledge within the reference public in the Benelux.

To this, the GC responded that:
  • to those who would not think of ‘Chiara’ as a female name, the word elements of either mark would be devoid of any conceptual meaning; as such, said elements could not be considered similar on a conceptual level;
  • those, who would instead think of ‘Chiara’ as a female name, would also think of ‘Chiara Ferragni’ as a name and surname.
Either way, just because the marks contain the same female name, this does not mean that the marks are conceptually similar:
indeed, the trade mark applied for identifies and distinguishes a certain person, that is a person belonging to the Ferragni family, while the earlier trade mark only refers to a name without also identify a certain person [para 68; again the translation is mine]
In the case at issue, the relevant consumer who was to recognize the term ‘Chiara’ as a name would not think that the earlier trade mark relates to products with the same commercial origin as Chiara Ferragni’s. 

It follows that no conceptual similarity would subsist between ‘CHIARA’ and ‘CHIARA FERRAGNI’. This conclusion is further reinforced by the presence, in the latter sign, of a figurative element (the eye) with a “very precise semantic content”.

As such, the Board had erred in considering that the conceptual similarity between the signs would be ‘neutral’.

A few Kats among Forbes's Top Influencers - Pets
Likelihood of confusion

Coming to the assessment of likelihood of confusion, the GC found that the erroneous conclusion - by the Board - concerning the visual similarity between the signs had also led to an erroneous finding concerning the subsistence of a likelihood of confusion. 

The GC concluded [para 86, again my translation] that:
despite the existence of an identity or similarity between the products at issue, the differences between the signs, in particular from a visual standpoint, are sufficient to exclude the subsistence of any risk of confusion in the perception of the relevant public.

This decision is a useful reminder of how, on the one hand, undertaking the assessment of similarities between signs is not an easy exercise and, on the other hand, one should really avoid applying automatically criteria and principles established in case law and office practice. 

From reading the judgment, the latter seems to be the main message of the GC: the criteria and principles established to assess similarities should not be intended as a checklist of points to 'tick', but rather as guidance that needs to be tailored and applied carefully in any given case, based on the relevant circumstances at issue.

[Originally published on The IPKat on 9 February 2019]

Tuesday, 5 February 2019

The value gap proposal in the latest Franco-German deal: what are the key points?

As readers who have been following the discourse around the draft Directive on Copyright in the Digital Single Market [Katposts herewill know, a few days ago the closed-door trilogue negotiations encountered a major obstacle, due to disagreement over one of the most debated provisions in this proposed piece of legislation: Article 13, also known as the 'value gap' or 'transfer of value' proposal [Katposts here]

While France has supported the idea that the obligations within Article 13 should be generally applicable, Germany has advanced the view that an exemption should be instead available to undertakings with annual turnovers below a certain threshold (EUR 20 million).

Concept of 'value gap'

Article 13 is a provision which, in the original intention of the EU Commission, would serve to remedy what has come to be known in jargon as the ‘value gap’ or ‘transfer of value’. 

This concept refers to a claimed mismatch between the value that some digital user-uploaded (UUC) content platforms are perceived as obtaining from protected content and the revenue returned to relevant rightholders. The argument for supporting that such a gap exists is the inconsistent application of the safe harbour for hosting providers under Article 14 of the E-commerce Directive

The consequence would be that, on the one hand, rightholders would not be appropriately compensated for the exploitation of content to which they own the rights and, on the other hand, competition in the market would be distorted.

The evolution of Article 13

In its original formulation, the value gap proposal provided that, where a hosting provider stores and gives public access to large amounts of user-uploaded works (the concept of ‘large amounts’ is however undefined in the proposal), thereby performing an act of communication to the public within the meaning of Article 3(1) of the InfoSoc Directive, it would be obliged to conclude a licensing agreement with the relevant rightholder(s), unless it is eligible for the safe harbour pursuant to Article 14 of the E-commerce Directive

Irrespective of whether the safe harbour in Article 14 of the E-commerce Directive applies, a provider that gives access to large amounts of works should take appropriate and proportionate measures to ensure protection of works or other subject matter, also by implementing effective technologies.

By means of a simplification, the original formulation of the value gap proposal was rooted within three core ideas:
  1. That the storing and making available of protected content uploaded by third parties would qualify as an act of communication to the public by the hosting provider itself;
  2. That, despite the making of restricted acts, the safe harbour protection within Article 14 of the E-commerce Directive would in principle remain available to the hosting provider.
  3. That hosting provider would be subjected to an obligation to prevent the making available of allegedly infringing content. Recital 39 suggests that this could be done in the context of a collaboration between hosting providers and rightholders, in that "rightholders should provide the necessary data to allow the services to identify their content and the services should be transparent towards rightholders with regard to the deployed technologies, to allow the assessment of their appropriateness."
A few months ago, The IPKat reported that an agreement had been reached within Coreper (the Council's permanent representatives committee) concerning the text of the Directive; last September also the European Parliament managed to adopt its own version of the directive to use as a basis for a negotiating mandate in the context of the trilogue.

Today's leak

Today Politico leaked the text of the compromise eventually achieved within Coreper further to the Franco-German disagreement, which also required a call between Macron and Merkel. The next Coreper meeting is currently scheduled to take place on 11 or 12 February.

As far as the value gap proposal is concerned (other points tackled in the new text are the press publishers' right, the principle of fair remuneration, revocation right and text and data mining), these are the main points as put forward by the current Romanian Presidency of the Council.

Softer liability regime for small companies and start-ups

Micro- and small sized enterprises with:
  • a turnover of less than EUR 10 million, and (these are cumulative conditions)
  • whose services have been available to the public in the EU for less than 3 years, 
should benefit from a softer liability regime. 

This entails that in the first 3 years of activity these companies will be only subject to notice-and-takedown obligations. 

However, if the average number of monthly unique visitors exceeds 5 million, then these subjects shall be under an additional obligation to demonstrate that they have made best efforts to prevent further uploads of the notified works and other protected subject matter for which the rightholders have provided relevant and necessary information.

Comment - From the text as put forward, it appears that the 'softer liability regime' would become inapplicable essentially in 3 cases:
  • Because of the enterprise's own growth (turnover)
  • Because of the enterprise's own popularity (visitors)
  • Because of the passing of time: if 3 years have passed, then it appears that the ordinary liability regime would start applying irrespective of growth. This suggests that, rather than a 'softer liability regime', what is envisaged here is a cooling-off period on the application of the ordinary regime.
Kat-fine print
Safe harbour availability

With regard to safe harbour availability, the Coreper text states that protection remains "possible" and that Article 13 should not be intended as imposing a general monitoring obligation. However, the safe harbour would be "for purposes falling outside the scope of this Directive" (Article 13(3)). 

Comment - This part seems aimed at those who have criticized the value gap proposal for failing to comply with the E-commerce Directive. While the new Coreper text reiterates the possibility for the safe harbour to apply, one should be careful: the language suggests in fact that the safe harbour would only potentially apply to types of liability other than copyright-related. 

In other words, if a platform communicates to the public (together with users of its services), then the safe harbour would not apply tout court

The situation might be potentially akin to the following: let's assume that a user uploads a video on a UUC platform that triggers their own liability, eg, under a certain Member State's criminal law (eg for defamation). That video also includes a song. The platform might in principle enjoy the safe harbour protection in relation to the defamatory activity of the user, but would not be able to do the same in relation to the making available of the protected song.

What happens in case a licence is not obtained?

Rightholders shall be under no obligation to grant licences. 

If a platform makes unlicensed content available it might be liable for unauthorized acts of communication to the public, unless it demonstrates that it:
  1. made best efforts to obtain a licence, and
  2. made best efforts to prevent the availability of infringing content in respect of which the rightholder has provided relevant and necessary information, and 
  3. acted expeditiously, upon receiving a sufficiently substantiated notice by the rightholder, to remove or disable access to that content, and 
  4. made best efforts to prevent the future availability thereof.
The concept of 'best efforts' entails an appreciation of whether the platform has acted as a 'diligent economic operator'.

Comment - If a rightholder does not want to grant a licence, they are acting in their full right. However, the platform should seek to obtain one and, failing to do so, it might still allow the availability of the unlicensed content, provided that it complies with the requirements above. This is what is called a 'mitigation of liability', likely aimed at mitigating the criticisms of those who dubbed Article 13 a 'censorship machine'. No censorship might occur after all [for a different take, see MEP Reda on her website; also critical, see CREATe].

What if also your Kat was actually BIG?
Users' rights

While the new text does not endorse the idea (flagged in the past) that a user-generated content exception be adopted, it reiterates that users shall be allowed to upload content generated by them and which includes third-party content for purposes of quotation, criticism, review, caricature, parody, or pastiche:
Users shall be allowed to upload and make available content generated by themselves or by other users and which includes parts of existing protected works and subject matter for the purposes of quotation, criticism, review, caricature, parody or pastiche.
Comment - Some commentators might consider this part of the text as reiterating the status quo, but that would be hardly correct.

If this language stays, it might have a major impact: in fact, all the concepts mentioned above (quotation, criticism, parody, etc) are currently framed in Article 5 of the InfoSoc Directive as optional exceptions which Member States are free or not to transpose into their own laws.

This new wording of Article 13(5) might make at least the exceptions in Article 5(3)(d) and (k) de facto mandatory for Member States to have.

Alternatively, Article 13(5) of the new directive might be considered as overriding Article 5(3)(d) and (k) of the InfoSoc Directive. If that was the case, it would be BIG (but would depend on whether the InfoSoc Directive is included among the pieces of legislation the new directive would be without prejudice to).

Let's now see how things develop ...

[Originally published on The IPKat on 5 February 2019]

Sunday, 3 February 2019

Writing a proposal for an LLM/PhD thesis: what do you need to think about?

UniKat: Working hard ...
not to fall asleep
Several students are currently busy choosing or will shortly choose topics to develop in the context of their own university theses and dissertations, including at the LLM/master level and in the context of PhD proposals/early stages of their PhDs.

Often, students have to submit a ‘Proposal’, which tackles how they intend to develop their work.

The allowed word count for a thesis/dissertation varies: it is approximately somewhere between 15,000 and 25,000 words for an LLM dissertation and 75,000-100,000 words for a PhD. 

The funny thing about LLM/PhD proposals is that one is expected to know what they are going to write about before they have done all the required research and actually know what needs to be written about.

Several books and articles have been written on how to develop a good piece of academic research, whether as a student or as a professional researcher. 

However, not much is available online (a notable and valuable exception is this) in a quick and accessible format that explains what to consider in the absolutely early days of one’s own brainstorming activities, when one might not have many ideas or, if they do, these are likely to be embryonic at best.

On top of all this, hearing words like ‘research questions’, ‘methodology’ does not really help, as the likely thing to happen is that a big question mark is stamped all over them. 

So what can be the absolutely initial steps that a student needs to consider when approaching the preparation of a proposal for an LLM/PhD dissertation, so that even the drafting of the proposal may feel a bit less of a daunting task than what it is? 

In my view, a student should ask themselves these 5 key questions: 
  1. What do you want to spend your time on for the next X months if not even years? 
  2. What is the key question that you wish to answer with your work? 
  3. What issues or sub-questions do you need to tackle/answer in order to answer your key question? 
  4. What should/will you leave out of the scope of your inquiry? 
  5. How do you go about answering your questions? 
What are these questions relevant for? 

Answering Question 1 

… allows you to choose a topic that is worth committing to: if you do not care about/are not remotely interested in that area at the outset, drop it. 

There is probably little that is more ill-conceived than choosing a topic one is not interested in, thinking that it would be good career-wise, that your professor would like it and you would receive a better mark … 

At this stage of your career, you can afford something that might become a rare luxury in the future or something that in certain other professions is not even conceivable, that is to focus on what interests you. 

Your job is to make it interesting for others, too. Let your passion be infectious, and enjoy undertaking your research. 

In other words, there is no topic that is off-limits, insofar as you have something original to say on it. 

Answering Question 2 

… allows you to think about something that in the formal context of your thesis proposal is the ‘Aim’ of your research. 

It also allows you to narrow the focus of your inquiry down to something more specific, which can be achieved during both the limited (hopefully) time that you have at your disposal as an LLM/PhD student and in the allowed word count. 

Answering Question 3 

... allows you to refine the scope of your inquiry even further and think rationally about the key question that you have identified in the previous step, in a way that somewhat resembles the procedure that one needs to follow to demonstrate a mathematical theorem. 

What do I mean here? 

When I was in high school in Arezzo (Tuscany, Italy) we had a mathematics professor, Prof Martelli, who was both well-respected and well-feared by students. He taught us coding (back in the late 1990s, when something like this was not yet cool or the ‘must-do’ thing) and every week he called us to the blackboard to demonstrate mathematical theorems, as explained in the textbook he had written. Most students found this exercise extremely daunting, and thought that learning the demonstration ‘by heart’ was the way to go. 

WRONG: you did not need to memorize anything, if you actually understood the relevant steps in the demonstration. It was an exercise of understanding, not a memorization exercise. 

This is an approach that is helpful, I think, in any other field, including legal research: your work should flow somewhat akin to a demonstration, in that all descriptive parts must serve to build your critical arguments. If there is a part that does not serve this overall goal, that part is probably useless or should be re-considered.

So, you need to ask yourself: what questions do I need to answer in order to answer my key question and fulfil the aim of your research? 

If you are able to identify 4-5 sub-questions which are logically linked to each other, not only will you have a sound structure for your research, but you will also have found your ... Table of Contents. Your dissertation can be structured on the basis of the sub-questions that you have identified this way. 

This will also allow your entire analysis to develop in a sound and logical fashion.

Answering Question 4 

… serves for you to set the ‘Limitations’ (or ‘Delimitations’) of your thesis: of course, you have to account for limited time and limited available word count. This means that you have think about an appropriate scope for your work, and explain why you are leaving out what might have been potentially relevant. 

This part will also serve you to identify future work that needs to be done in this area – whether by you or others. 

This way, you will actually be a voice in the broader community of researchers active in the area you have chosen. 

Answering Question 5 

… serves you think about your method. It is nothing mysterious: like in every other problem-solving scenario, you need to think not just about what you need to do, but how you are going to do it. 

If one of your sub-questions requires you to consider the approach in another jurisdiction, then you will likely need to employ a comparative method; if, instead, one of your sub-questions requires you to appreciate a certain aspect in light of the rationale for protection of right XY, then you would likely need to engage with theoretical methods, etc. 


Thinking about the 5 questions outlined above would allow you to draft a proposal that contains: 
  • The aim of your research 
  • Your research framework/questions 
  • Your table of contents 
  • The limitations to your research (and indications for future work) 
  • Your methodology 
… and write something that you enjoy doing: you have a lot of freedom, so use it wisely.

Other IP student materials that I have developed are available here

Any feedback/comments from IPKat readers on their experience as researchers or students will be also gratefully received.

[Originally published on The IPKat on 3 February 2019]