Vagelis Papakonstantinou* and Paul de Hert**

 Abstract

Museums are, in most cases, publicly-owned holders of vast amounts of information that are, by definition, open to everyone. Location restrictions, however, usually limit public access. The Internet could change this: once museums digitise their collections and upload them onto their Internet sites, online access would be possible for anyone, anywhere. The difficulty in this case would be that there are practically thousands of museums around the globe, ideally each maintaining its own Internet site. Users therefore face substantial difficulties when conducting research online. From this point of view it is probably a self-evident development to aggregate online museum content in a single website, in order to facilitate user access. This explains the initiatives, for instance, of Europeana from the public sector and the Google Art Project from the private sector – each one in terms of content volume and user exposure holds a pre-eminent position among its (Internet) peers. These initiatives, however, are disruptive, both as regards business methods and legal systems, challenging traditional notions and treading at the borders of well-established legal principles and long-serving rules and regulations. This article discusses the legal issues raised by the contemporary aggregation initiatives of museum content over the Internet, by reference to the above two initiatives. Questions relating to copyright, the sui generis database right, as well as, the issues of systems’ interoperability, public sector information and restitution will be addressed in the analysis that follows.

Cite as: V Papakonstantinou and P de Hert, “Legal Challenges Posed by Online Aggregation of Museum Content: The Cases Of Europeana and the Google Art Project”, (2012) 9:3 SCRIPTed 314 http://script-ed.org/?p=713

Download  options

DOI: 10.2966/scrip.090312.314

 © Vagelis Papakonstantinou and Paul de Hert 2012.

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 2.5 UK: Scotland License.

1. Introduction

Internet trends, ultimately constituting social phenomena, change over time. The ways providers make their data available online and users navigate the web alter frequently. Allegedly, for the time being we are living in the age of cloud computing, web social networking, micro-blogging, and, what is of direct interest to this article, heroic Internet entrepreneurship. The latter may be identified as bold, new Internet projects of infinite space for development, uncertain outcome and even more uncertain profitability and, thus, sustainability. These projects are aimed at nothing less than affecting the daily lives of everyone. They are advocating a worthy cause but at times they may carry hidden financial considerations. Examples of such projects relevant for this article are, from the private sector, Google Books[1] and the Google Art Project,[2] from the public sector, Europeana,[3] and from non-profit institutions, Project Gutenberg[4] or Internet Archive.[5]

Museums are, perhaps unavoidably, entangled in these projects. They are in most cases public-owned holders of vast amounts of information that, by definition, is open to everyone. However, location restrictions limit public access; at best, the trend for touring exhibitions and loan of artifacts, as burdened with a series of factual and legal difficulties, compensates for location limitations. However, the Internet could change all this: Once museums digitise their collections and upload them onto their Internet sites, online access would be possible for anyone anywhere.

The difficulty in this case would be that there are practically thousands of museums around the globe, ideally each maintaining its own Internet site. Works of art are dispersed among them without any logical coherence. Users therefore face substantial difficulties when searching online for specific works of art or conducting research. From this point of view, it is probably a self-evident development to aggregate online museum content to a single website, in order to facilitate user access. Users shall only need to log into a single website rather than helplessly navigating the web in order to assemble a complete picture of an era or an artist, being compelled to visit the websites of a number of museums in the process. This explains the heroic Internet entrepreneurship behind the initiatives of Europeana, from the public sector, and the Google Art Project, from the private sector.

What is also common among all heroic Internet entrepreneurship is that it is unavoidably disruptive, both as regards business methods and legal systems. For instance, ambitious projects such as Google Books have overturned long established business practices and have faced substantial legal difficulties in the process.[6] Other similar projects, in an effort to avoid legal controversy as much as possible, have proceeded with extreme caution. Project Gutenberg, for instance, makes available only acclaimed works with expired copyright.[7]

The aggregation of digitised (or digital) museum content onto a single website threatens to undermine the traditional scheme for the use of museum collections. Until the relevant initiatives came into light, museums made their collections available to the public either in the real world, from within their premises, or through their own websites. The connection, therefore, between a museum and its collections was ever-present and predominant. This essentially meant that all decisions on the use of artifacts lay within the museum: museum managements were free to decide which collections to display from time to time, which to place on their respective Internet sites, what to charge for each category of uses, as well as, the terms and conditions regulating such public access or re-use. Granting aggregation rights to a single website operator could mean that certain of the above decisions no longer are the prerogative of the museums concerned. This shift in the decision-making process is important and bears concrete legal consequences for museums.

Online content aggregation probably constitutes a worthy cause, as public access and openness are perhaps better served through single, all-encompassing websites rather than dispersed and incompatible among the dedicated museum Internet sites. As it will be demonstrated in the analysis that follows these conflicting interests are, for the time being, addressed only in the texts of the relevant aggregation agreements. It is a solution, however, that lacks in transparency and legal certainty and therefore needs to be replaced by concrete regulations.

In addition, online aggregation of museum content per se affects the current public access model and therefore, unavoidably, the legal systems supporting it. From this point of view, the copyright system, as well as the EU-particular database right, is evidently affected. The same applies to user expectations to system interoperability, as well as, to public access and re-use of public information. Does online content aggregation constitute an intended act of re-use and exploitation of public sector information in a more efficient and financially meaningful way, as expected by the relevant EU legislation?

Finally, although the right of museums to grant the necessary powers to online content aggregators is taken for granted for the purposes of this analysis, what should not be taken for granted is that online content aggregation, once in place, is easy to place within contractual or technical boundaries and even becomes possible to predict. As it will be demonstrated, online aggregators, even under their best intentions, cannot possibly promise a non-profit (or, for the same purposes, profitable) use of museum collections on their websites. Once aggregated on-line, museum content develops an independent existence, potentially exceeding the limitations and purposes, of the museums that released it.

The foregoing remarks and questions, learn that initiatives for the aggregation online of museum content are bound to open legal discussions. A number of traditional notions, well-established principles, and long-serving rules and regulations are challenged during the execution of what would otherwise appear to be a necessary Internet process to facilitate user access: superimposing an aggregating website upon content scattered online or stored in off-line databases.

It is exactly these challenges posed by online museum content aggregation to existing legal schemes that will be elaborated in this article, with the aim of highlighting shortcomings of the regulatory framework in effect while accommodating the needs of a previously unforeseen, but by now essential, content use model. In order to achieve this, two concrete models shall be placed under scrutiny (under section 1): the authors consider that the contemporary initiatives from Google (the Google Art Project) and the EU Commission (Europeana) constitute an adequate sample due to their shared purpose but otherwise substantial differences in origin, approach and implementation.[8]

Both initiatives make content available through the Internet. Both initiatives aggregate museum content in a single Internet space, although each one has implemented a different approach and strategy and each presents significant advantages but also challenges to museums. In this context, questions relating to copyright applicability (3), the sui generis database right (4), as well as, the issues of systems’ interoperability (5) public sector information (6) and restitution (7) will be addressed in the analysis that follows. The analysis that follows shall demonstrate the limitations of the regulations currently in effect while addressing the new, online content aggregation, reality, and the need for an ad hoc regulatory framework in order for online museum content aggregation  to develop its full potential to serve user access.

2. How do Europeana and Google Art work?

2.1 The birth of the Europeana and Google Art initiative

On 20 November 2008, Europeana.eu was formally launched by the President of the European Commission.[9] The political initiative behind its establishment may be traced as early as in 2005,[10] allegedly triggered by the launch of Google Books. The aim of the Europeana project, although originally connected to the EU Digital Libraries Initiative, was “to make Europe’s cultural, audiovisual and scientific heritage accessible to all”, exceeding thus the limitations of a book digitisation project. By now, Europeana’s strategic plan is to provide access to Europe’s entire cultural heritage, intended to be digitised by 2025.[11] The project gave access at the time of its launch to 4.5 million digital objects from over 1,000 contributing collections from every member of the European Union, which increased to 19 million objects by late 2011. Perhaps true to its bureaucratic origins, the Europeana model is multi-layered: At the highest level, a foundation, the Europeana Foundation, is the project’s governing body. The Foundation comprises an Executive Committee, a Board of Participants, and the Council of Content Providers and Aggregators. At national level, the Europeana project is ideally intended to co-operate with a single content aggregator per Member State. However, if this is not possible, as has often been the case until today, Europeana may contract directly with cultural institutions.

Google launched its, equally ambitious but seemingly far less bureaucratic, Art Project on February 1, 2011. The Project, apparently an off-spring of the famous Google employee-policy of 20% project,[12] is an online compilation of digitised museum content (paintings) from galleries worldwide as well as, making indoor use of the Google Street View technology, a virtual tour of the respective museums. In this way, at least as per its mid-2012 structure, users are provided with two options: either to view single artworks, admittedly in high-resolution normally inaccessible to museum visitors, or to virtually walk through the co-operating museum collections. At the launch of the Google Art Project seventeen galleries and museums were included, from both sides of the Atlantic, and by mid-2012 this number increased to more than 150.

As promised, we will focus in this legal study on these two relevant models dominating the Internet at the time being.[13] Both Europeana and the Google Art Project share a common purpose, to create a single website where users may access the whole, or a very substantial part, of human cultural heritance. On the other hand, due to their difference in origins (public vs. private), approach (Europeana has adopted an all-inclusive approach, aiming at including any and all European culture objects, whereby the Google Art Project provided access at first only to paintings and then to other types of collections as well however from a limited number of museums) and implementation (Europeana does not host any content on its website, while the Google Art Project hosts both museum content and navigation details), the above two models constitute an adequate sample through which to elaborate upon the legal issues raised either individually or in common by those two models. Given the nature of the initiative, it is to be expected that no more than a handful of similar initiatives around the world may exist at any given moment.[14]

2.2 The digitisation of museum content (step 1)

The digitisation of museum collections is by now a self-evident task: it enhances not only public access, if the digital files are subsequently placed online, but also their preservation, as well as, if applicable, their commercial exploitation. For the time being such digitisation essentially includes the digital photography of artifacts accompanied by the possibly extensive tagging (adding of information) of the resulting files.[15] Future technologies may enable additional options, such as 3D printing. At any event, the result of the digitisation process is, practically, one or more digital files per artifact, complete with the relevant information (metadata). Once assembled, these files may be treated as usual computer files; uses may vary ranging from mere storage for preservation purposes to further processing in order for the files to be uploaded onto websites or be made available to the public or the market.

Although an established priority by now, the digitisation of museum content is by no means a straightforward process. A number of considerations, admittedly mostly of a technical nature,[16] need to be made and numerous strategic assumptions need to be re-evaluated constantly[17], given not only the technology options available from time to time, but also limitations in resources. However, in itself it is a process well under way, accepted as given wisdom by most museum managements around the world.

From a legal point of view, difficulties arise as soon as the digitisation process begins. As noted, a ‘digitised’ artifact consists today of digital photographs and tag computer files. With regard to these photographs, it remains yet unclear whether they constitute ‘works’ under the copyright regime; in addition, even if copyright was acknowledged on such photographs, it is doubtful whether they should benefit from full, renewed copyright protection or whether certain open-access limitations must be conceded, particularly when pertaining to out-of-copyright artifacts[18]. On the other hand, metadata files will normally be proprietary to the museums that created them.

Apart from a lack of clarity as to the applicable regulatory framework, another legal issue relates to ownership of (any proprietary, if applicable) rights. Normally, museums are expected to own the rights over their digitised collections. However, different countries have established different policies as to the proprietary rights of museums onto their collections. Although clear and unencumbered ownership must not be excluded (particularly in the cases of relatively ‘recent’ collections), it is also possible that ultimately it is the state concerned that is the owner of all or certain museum collections (as is usually the case in archaeological or history museums). Further complicating things, it is also possible that artifacts displayed in museums actually belong to third (private) parties. In addition, the legal nature of museums may differ considerably, ranging from public institutions to semi-public or entirely private organisations, public or private foundations etc. Any generalization is therefore impossible.[19] While the case is normally that a certain museum is free to decide to digitise its collections, and it evidently owns the rights over the resulting files, it is also entirely possible that another museum must first acquire a state (or, a private party) permission for the same process and the resulting files are not entirely its own.

Notwithstanding the above, for the purposes of this article, unless otherwise explicitly stated, it shall be assumed that museums, regardless of ownership rights over their actual collections, do own the (copy)rights over the relevant digitised files and that they are also free to dispose them in whichever way they like.[20]

2.3 Making available museum content (step 2)

Once they have digitised their collections museum managements are expected to make available part or all of them on the Internet. Nevertheless, even if this is actually achieved, still the average Internet user may find himself helpless when searching for an artifact or researching on an era. As known, museum collections are rarely exhaustive of an era or even of a single artist; artifacts are normally dispersed around the globe without any coherent connection as to their whereabouts. In fact, quite the contrary is often the case, given not only museum (and the respective countries’) antagonisms over time, but also the dislocation of artifacts that occurred more often than one would like to admit in the past.

This is why the online aggregation of museum content in a single Internet space, whereby significant parts of human cultural heritage would be accessible to everyone, is at the time being a mainstream Internet trend, as illustrated by the two initiatives discussed in this article, Europeana and the Google Art Project. Each one has implemented a different approach and strategy and each presents significant advantages but also challenges to museums.

2.4 How does Europeana make museum content available?

Europeana, an option admittedly available only to European museums, is, perhaps, more of a dedicated search engine than a research website per se.[21] This is perhaps inevitable, given the breadth, and width, of its contents: It includes anything from books and texts, to digitised museum artifacts, audio files, video, photography, etc. Content providers also vary considerably, ranging from museums and libraries, to public and private foundations, collections, publishers, etc. Museum content is therefore only one category among the many that are available in the Europeana content aggregating website. Evidently, this affects several aspects of content aggregation, among which the content display, its indexing, the search facilities, as well as, inevitably, the relationship between the museum concerned and the aggregating website operator (in this case, the Europeana Foundation).

As per the relevant Data Exchange Agreement,[22] Europeana ‘data providers’ (that is, museums, at least as far as this article is concerned) are to submit to its website only ‘previews’ and ‘metadata’ of their digitised collections. This essentially means that Europeana uploads, and makes available to its users, only these two items (practically, thumbnail images and a limited number of information on each artifact) and not detailed digitised files. Europeana itself does not host in its servers any content, at least for the time being.[23] Visitors to the Europeana website, once they identify through its search facility the items that interest them, are subsequently guided through hyperlinks to the museum website concerned.

Consequently, as far as the Europeana model is concerned, (European) museums only need to upload onto the relevant website by-products of their, supposedly already under way, internal content digitisation process: reduced quality thumbnail images of their digitised collections, as well as, all or part (depending on the interoperability rules) of the relevant metadata. It is up to museums to decide which items to show in Europeana – no general obligation to upload their entire collections exists to-date. Europeana has also adopted an all-inclusive and open approach, whereby any and all content providers in Europe are encouraged to upload content onto its website.

Interoperability among systems is warranted through a set of guidelines prepared by Europeana that need to be followed by the participating content providers, as per the Data Exchange Agreement.[24] Copyright issues, as will be later explained (under section 3) are referenced to the content provider concerned. Finally, no term on exclusivity is included in the same agreement; museums are free to participate in Europeana and other similar projects.

2.5 How does Google Art make museum content available?

The Google Art Project has implemented a more simple (but perhaps less ambitious) strategy compared to that of Europeana. It only co-operates with museums, which, whenever selected by Google to enter its content aggregating website, are apparently asked for two things: highly digitised files of their collections and access to their premises in order to use its Street View technology. It is up to museums to decide which artworks to display in the relevant website, and to decide which part of their premises to provide access for the walk-through purposes. Users in the Google Art Project website have two options: They can either navigate through the respective museum corridors, or they can view individual artworks through high quality images that are, apparently, hosted on the same website. No links to the museum websites are available: once in the Google Art Project website, users stay in throughout their research.

In the case of the Google Art Project the agreement entered with the museums concerned has not been made public, a persistent Google policy[25] that ought perhaps to be kept in mind when assessing the merits of each model. It is therefore unclear whether the same agreement is entered with each contracting museum, or whether special terms and conditions are agreed, for instance on the co-operation duration and options for (and after) termination, on exclusivity (for the same high-resolution images or for navigational information), or even on any future income distribution scheme. Although copyright and user rights issues will be discussed later (in section 3), here it is only noted that a distinction needs to be made between the images provided by museums and the Street View navigation material: the former are owned by the museums concerned, while Google is the rightsholder of its Street View material.[26]

Interoperability is not an issue in the Google Art Project case, since Google aims to create a framework whereby users stay within its website limits, and use only the material available therein, while the limited number of museums and the closed system architecture essentially means that Google itself takes care of the proper upload and placement of material onto its website. Finally, as is the case with Europeana, this is a non-exclusive option for the participating museums which may choose to provide content to other online content aggregators as well.

3. The legal nature of aggregating museum content online

The aggregation of museum content online is the end of a process that essentially involves three stages: the digitisation per se of museum collections, the placement of (some of) the resulting digitised files onto the museum’s own website, and, finally, the making available of such digital files to third parties acting as online content aggregators, in this case, to the Europeana Foundation and Google respectively.[27]

Once created, the resulting digital files are expected, under the distinctions made above,[28] to constitute a ‘work’ by themselves from a copyright point of view, to which the rightsholder is the legal entity undertaking such digitisation. Evidently, any conditions of the digitisation permit will follow, limiting, their subsequent uses.  For example, if the digitisation of a statue or a painting was explicitly allowed for preservation purposes, the lawfulness of its subsequent use in content aggregation websites ought not be considered self-evident.

The placing of such digitised material onto the respective museums’ websites, if the latter are perceived as extensions into cyberspace of the museums’ existence in the real world, most likely shall not exceed the relevant digitisation permits, because it can always be held that such uploading is in fact nothing more than the performance into cyberspace of real world-permitted re-use actions. Therefore, always depending on the actual wording of the relevant agreements, this type of uploading is not expected to require a renewed rights clearance with regard to the digitisation permits concerned.

However, online content aggregation onto the websites of third parties, regardless of whether public or private, would not seem, prima facie, to be covered by the original digitisation permit, unless of course otherwise stipulated therein. Although such aggregation may still serve cultural and public access purposes, the fact remains that a re-use of the digital files takes place in one way or another (either reduced into thumbnail images with metadata or in their best digital analysis) – in essence, a sublicensing to a third party. Such re-use practically exits museum natural or cyberspace boundaries – the fact that the Europeana website links to the museum website does not affect much this assertion. From this point of view, it appears that a digitisation permit that supposedly grants to the museum a right to create digital files of the artifact for preservation and public access purposes may not easily be perceived as implicitly allowing to the same museum a sublicensing right to online content aggregators for further use of the same files.

The above distinction may appear legalistic, especially in view of the noble causes served by the, admittedly non-profit, nature of the initiatives discussed in this article. However, in those cases where a museum is not the rightsholder of all rights over an artifact within its collections, and a digitisation permit was thus obtained by a third party, it merits closer examination whether such permit freely allows for online content aggregation sublicensing (particularly given that it is museums that transmit the digitised files to the Europeana Foundation or to Google, and not the latter taking pictures and then placing them online).

Once again, however, it is clarified that for the purposes of this article it is assumed that museums are indeed the sole rightsholders over their collections, and that they have the right to do with them as they please.[29]

4. Copyright and online museum content aggregation: who owns the content, and what can users do with it?

4.1 General principles

Despite the fact that a variety of rules may follow each artifact within a museum collection, the need in a content aggregation website is for a uniform and possibly simple copyright model. Online content aggregation would not be possible if the respective website had to translate a variety of copyright limitations into webpage access rules. On the other hand, an overly restrictive content aggregating website practically has no reason for existence – and will ultimately be condemned by a low number of user visits. This is why both models discussed in this article tread very carefully with regard to copyright issues, attempting to achieve a balance that will reassure museums and rightsholders without at the same time scaring users away.

The relationship between copyright[30] and online museum content aggregation, despite the simplifying efforts of the content aggregators in the two models under consideration, may prove complex and far from straightforward. Two sets of questions arise in particular: from the user perspective, the copyright-related questions refer to who owns the content displayed in aggregating websites and what uses over it are permitted each time. From the museum perspective, the same questions refer to the relationship between museums and content aggregators – essentially to the terms and conditions of the relevant copyright licensing agreement.

As far as users are concerned, Europeana, being more of a search engine than a direct source research tool, ultimately refers their copyright-related questions to each individual content provider. In principle, the terms of service of the Europeana website permit free use of all of its content except from the thumbnail images (and other previews), whereby the conditions applicable may be found alongside each individual image.[31] This may strike users as odd, given that it could be claimed that the added value of Europeana is exactly these previews or thumbnail images. However, it appears that the variety of materials assembled within the same website as well as the multitude of content providers is such that a uniform copyright solution would be impossible to implement. This is why users may come across in-copyright and out-of-copyright material, as well as a number of in-between variations and restrictions, while navigating the Europeana webpages. In order to lawfully process such material, or for requests for high(er) resolution images or for commercial uses of the same material, users are advised to consult each time with the providing institution copyright policy.

Therefore, in practice the Europeana model addresses the copyright question in a simple and straightforward, but perhaps also unambitious, way.[32]As such, the Europeana model could be accused of perhaps making limited use of its legal options at hand. In particular, Directive 2001/29/EC on the harmonisation of certain aspects of copyright and related rights in the information society stipulates in its Articles 5.2 and 5.5 that Member States may provide for exceptions or limitations in respect of specific acts of reproduction by publicly accessible libraries or by archives, where they are not for direct or indirect economic or commercial gain, when such exceptions and limitations do not conflict with a normal exploitation of the work or other subject-matter and do not unreasonably prejudice the legitimate interest of the rightsholder.[33] Given that the Europeana model probably meets the above conditions, it could be claimed that more could be requested by the Europeana Foundation (and the Commission) by Member State providers when providing content to the Europeana website.[34]

The Google Art Project apparently addresses the copyright questions of users, again, in a more efficient way than Europeana (admittedly, also assisted by its limited subject-matter). With regard to the relevant website, reference is made to the general Google Terms of Service.[35] There, it is clarified that “content presented to you as part of the Services…may be protected by intellectual property rights which are owned by the sponsors or advertisers who provide that Content to Google”. More specifically, the Terms of Service state, “you may not modify, rent, lease, loan, sell, distribute or create derivative works based on this Content (either in whole or in part) unless you have been specifically told that you may do so by Google or by the owners of that Content, in a separate agreement”. Consequently, users are allowed to work within the website, both on the high-definition images and on the navigation data, but they cannot amend or interact with them in any way.[36]

4.2 Resolving copyright issues through agreements

As far as museums are concerned, the copyright aspects of their relationship with online content aggregators are addressed in the relevant signed documents – essentially copyright licensing agreements.

Europeana, having published online its Data Exchange Agreement, constitutes the open and transparent option. The Agreement itself is a straightforward document, where it is stated that museums are free to decide which content to provide to the Europeana website, while at the same time maintaining their intellectual property rights for it, thus granting, in effect, a mere reproduction right.[37] The Agreement appears to be more occupied with interoperability and other functional issues for the website, than with content sharing details.[38] Overall, it appears to be a fair and balanced document, both as regards its terms and conditions and as regards the fact of its intended uniform use, meaning that more recognizable museums or collections do not get a better deal than others.

Such transparency is missing from the Google Art Project; as noted, the relevant agreement(s) entered with museums are not made available online. As such, their content may only be inferred from the actual operation of the relevant website. It therefore appears that museums grant their own files to Google to upload and host itself. Ownership remains with museums. The same agreement obviously grants access rights to Google Street View technology to navigate through the museum concerned. Museums are listed in the website without any alphabetical or other, easy to trace, method, a fact to-date not causing much access problems given their limited number although this fact may very well change in the future.

The lack of access to the licensing agreement between museums and online content aggregators ought not to be taken lightly. Because it is ultimately a contractual relationship (no particular legislation exists at this present time), the parties are free to regulate it in any way they see fit. Therefore, the particular terms and conditions included therein do matter. Given also that these agreements are most likely expected to be, in one way or another, formally (as in Europeana) or commercially (if the Google Art Project takes off) non-negotiable, since museums will normally wish to enter the content aggregating model and not vice versa,[39] the relationship risks turning from a co-operation between equal parties to a conflict between adversaries.

A series of important issues are customarily discussed in such content licensing agreements. For instance, is there an obligation for museums to upload content or to continue uploading periodically new content in the aggregating websites? Or, are they obliged to upload their most popular content, or any content they like? From the aggregators perspective, is fairness in display to be evidenced by listing all museums and all content alike? Or, are they allowed to use the more popular artifacts in order to promote their own page views by users? Exclusivity is also an issue: are museums free to participate in any number of aggregating initiatives – perhaps, providing access to the same content to multiple players?

Another important issue is rights infringement: if any intellectual property rights are infringed through the content aggregating website (either by users misusing their access rights or by the museum itself, having uploaded content without proper authorisation), what are the policies applicable each time by the content aggregator?

The above issues may seem secondary, particularly in view of the noble causes served by the relevant initiatives, but they may prove to be very important in practice, ultimately turning a win-win relationship for museums, content aggregators and users alike, to a relationship of conflict and a source of friction. This is why clarity and transparency is necessary, and, given also the public benefit character involved, it becomes a condition perhaps met only in one of the models examined in this article.

5. The impact of database rights

5.1 General principles

Database rights, created in the 1990s as a response to the, real or perceived, global rise in the database sector,[40] protect the database maker instead of its author, despite of the fact that it is examined within the intellectual property rights field.  Introduced fifteen years ago, these ‘newly recognized rights, remain popular.[41] In this article we are interested in understanding the way these rights interact (or, do not interact) with online museum content aggregation. First, with regard to the museums themselves, as content providers: do they have a sui generis database right over their digitised collections? Second, from the online content aggregators point of view: having gained re-use rights for museum digitised content, do they themselves acquire sui generis rights over their websites’ databases? And, third, from the users’ perspective: does existence of this right warrant additional uses and does it open up new possibilities?

Before unpacking these questions, a brief analysis should go to the way in which the EU database right operates. It should be noted that this right is not only unique to the  EU but, attention should also be given to the wording of the relevant regulations, as it has been substantially affected by important case law. Hence, the maker of a database is granted a sui generis database right only if he can prove that he spent substantial resources in the actual “obtaining, verification or presentation” of its contents – not simply if he happens to be the owner of a set of data that he then organized into a simple database.[42] This case law finding greatly affected public sector owners of large sets of data that were forced to open up their data resources to competing newcomers. In the same way, however, it may well affect museums, when they digitise and grant re-use rights over their collections to online content aggregators.

Another interesting aspect of the database right refers to the fact that it aims at separating itself from (any) copyright protection. In particular, a database may very well qualify for sui generis right on top of copyright protection[43]. In addition, what is perhaps even more important for the purposes of this article is that, if the contents of a database qualify for their own copyright protection, they are separately protected in their own right; their protection does not prejudice the protection provided also by the sui generis right[44].

5.2 Do museums have a sui generis database right over their digitised collections?

As far as museums are concerned, the end result of their, expensive and time-consuming, digitising efforts for their collections will most likely qualify both the database (“a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means”, Art. 1.2) and the substantial investment (“the maker of a database which shows that qualitatively and/or quantitatively a substantial investment in either the obtaining, verification or presentation of the contents”, Art. 7.1) criteria[45]. Museums are thus normally expected to be the rightsholders of a sui generis database right over their digitised collections, as organized into a database.[46] In addition, Public-Private-Partnership (PPP) schemes and other outsourcing options notwithstanding, given also that the Directive protects the maker and not the actual author, it is likely (and shall be considered as such for the purposes of this article) that the museum is the sole rightsholders of a database right over its digitised collections.

Once ascertained, the database right affords museums the right to permit ‘extraction’ and/or ‘re-utilization’ of the relevant files. The second is of particular relevance in this case, because it includes “any form of making available to the public all or a substantial part of the contents of a database by the distribution of copies, by renting, by on-line or other forms of transmission”. Consequently, even the uploading onto their own website is an act of re-utilization for museums. When making content available to third parties (online content aggregators), in order for the latter to make available onto their own websites, the case will most likely be for a, permitted, “contractual” “transfer, assignment or granting” of such right.[47]

5.3 Do the online content aggregators have database rights?

From the online content aggregators’ perspective, the sui generis database right affects them twofold: first, as licensees and, second, as licensors. As far as rights acquisition is concerned, as seen, museums have the right to transfer their sui generis right to third parties by means of a contractual license. Supposedly, explicit mention should be included in such agreement on this transfer; a broad reference to any and all intellectual property rights over the digitised collections under transfer would not suffice. In this context, the Europeana agreement indeed asks for content providers to “grant Europeana Foundation a non-exclusive, worldwide, royalty free license on copyright, related rights and the sui generis database right for the duration of this agreement” (Art. 2.2).

Once the museums’ digitised collections are uploaded onto the aggregating website, this will in turn constitute a database[48]. It therefore remains to be examined whether sui generis database rights are granted to its rightsholders (the online content aggregators) as well.

In this case the requirement for a substantial investment in the “obtaining, verification or presentation” of the database contents will apply. In other words, in order for online content aggregators to qualify for the sui generis right, they need to prove that they spent substantial resources, either qualitatively or quantitatively, either in the obtaining, verifying, or presenting of their websites’ contents. This can by no means be assessed in this article. As known, the Europeana Foundation does not pay museums for the uploading of their collections onto its website. On the other hand, it does spend substantial resources in maintaining its website (“presentation”), although not as much as one would expect, given that it does not host content and it uses open source tools. In addition, the money spent by the Commission for museums to digitise their collections could (or could not) contribute towards awarding Europeana, the Commission’s official website for these purposes, with sui generis database right protection.

Assessing whether the Google Art Project spends substantial resources in obtaining, verifying or presenting its data is an even more difficult task. The agreement it enters with museums is not known, therefore we do not know whether any money is paid for its services or not[49]. Neither is known how much money is spent on maintaining its website, or using its, already developed for other purposes, Street View technology to provide navigational information. Consequently, it is practically impossible to decide whether it qualifies or not for sui generis database right protection.

The above missing information is critical when assessing whether online museum content aggregators qualify themselves for the sui generis database right protection. If they do, then they are afforded with an additional layer of protection – particularly with regard to any out-of-copyright works they may exhibit. At any event, the Google Art Project shall also have to address the difficulty of transcending jurisdictions. The sui generis database right is EU-particular, but the Google Art Project is international. Museums from multiple jurisdictions appear on its website. And so, the exact conditions under which the sui generis database right may be invoked, either by Google or by users, are unclear.

5.4 The users’ point of view

From the user point of view, ascertaining sui generis right protection to online content aggregators may affect the permitted uses over content displayed in the relevant websites. In particular, the database right permits “insubstantial” use of such database contents for whichever purposes, provided of course that such use does not “conflict with a normal exploitation of that database” or “unreasonably prejudices the legitimate interests of the maker of the database”.[50] Other listed exemptions include use of a database contents for the purposes of “illustration for teaching or scientific research” or “for extraction and/or re-utilization for the purposes of public security or an administrative or judicial procedure”.[51] These actions may pertain to “substantial” parts of a database and may be performed without the authorisation of the database maker. Such user rights, that cannot be contractually overridden, may prove of importance particularly for the more closed environment of the Google Art Project.

Evidently, all the above have to be assessed also in relation to any protection over a digitised artifact per se. As explained, the sui generis database right does not affect any copyrights over each individual item comprising the contents of a database. If therefore, as seen above under sections 2 and 3, full copyright protection is acknowledged over the digitised files of museum artifacts, then the above permissions for users may not apply. On the other hand, it could be argued that online aggregation (database making) of items each burdened with different copyright statuses actively prohibits permitted uses within the sui generis right framework.: For instance, a teacher shall not be able to use a collection of material from Europeana over an artist for teaching purposes, if each single item comprising the collection has very different copyright permissions. From this point of view, a conflict between the two rights (over EU soil) perhaps appears inevitable, once online museum content aggregation becomes the norm, and the public grows accustomed to using the relevant aggregating websites in order to access museum contents online.

6. A user expectation to interoperability among applications?

Despite numerous efforts that span well over the last twenty years[52], to-date there is no global or even EU individual right (or industry obligation) to interoperability among Internet applications or even computer hardware and software. In fact, quite the contrary is the case: closed system architecture is actively used by Web 2.0 providers and software developers alike, particularly when releasing new technologies, in order to fence users in. Once users are lured into and grow accustomed to using new applications and technologies, they then realize that a switch among competitive providers and products is practically impossible or requires a disproportionate effort on their behalf; a seamless flow of information is usually not permitted by the system[53]. Because this is still a basic trend in Web entrepreneurship, it is expected to affect online museum content aggregation, taking also into consideration the heroic entrepreneurship involved and the limited sustainability potential in the market.

In the past, system interoperability was warranted through application of a variety of legal bases, including unfair competition and consumer protection, and also heavy political pressure, as was the case in Apple’s iTunes/iPod relationship. These days, a right to interoperability among Web 2.0 social networking websites is being debated in the data protection context.

In any event, a right to system interoperability would first and foremost enable users to easily switch among competitive providers, taking all their data with them. In the Web 2.0 environment this would practically mean that users are allowed to take their profiles, content, and personal information with them when switching between social network websites or other web platforms. Because in late 2011 both content aggregation models examined in this article allow, and indeed encourage, users to create their own profiles and exchange (cultural) information, they both fall under the general Web 2.0 interoperability debate.

In practice, the issue of interoperability affects online museum content aggregators twofold: first, as a concern and second as a competitive advantage. The concern refers to making the various museum collections interoperable among them onto the same website. Once however this is resolved, the competitive advantage takes form: Depending on how the content aggregating website is built, users may be forced to stay within its limits if they wish to fully enjoy its benefits, not being able to transfer their accumulated work (for instance, profiles, collections, etc.) elsewhere.

From their part, users have exactly the opposite interests: they wish to be able to move from application to application seamlessly, not having to retype and rebuild all their information as well as, evidently, wanting to leave no trace behind if the users so wish.

With regard to the contents acquisition or uploading stage, each one of the two models examined in this article each has implemented its own strategy. Europeana, given also the multitude of contents intended to be presented onto its website, has published a set of guidelines for content providers (museums), describing exactly how content is to be formatted, in order for it to be accessible on the website.[54] The uploading of content, compliant to these prescriptions, is a contractual obligation (through the Agreement) of content providers. One must also keep in mind that Europeana does not host content itself, but rather presents only thumbnail images and (certain) metadata information.

The Google Art Project on the other hand, as is to be expected, has not made any information known on this issue: given that it is museums that provide the high-analysis paintings’ images, it is to be assumed that Google itself undertakes the uploading onto its website, thus warranting the interoperability among content. – It is also important to note that, to-date, only a limited number of museums appear on the website. In addition, the navigational information is provided by Google itself, through its Street View technology, therefore no interoperability issues are expected from this point of view.

What is perhaps of more importance for the purposes of this article, since it affects user rights, is the interoperability among online content aggregating platforms. This would mean, in practice, that Europeana and Google Art Project users would be able to move easily between these two platforms, integrating their respective profiles, saved searches, downloaded or self-created content, etc. In addition, given Europeana’s all-catching scope, this would also mean that Google Books, another Google heroic entrepreneurship initiative, would be accessible to users of Europeana.

No such user right, or even online content aggregator obligation exists currently. This is perhaps to be expected by the Google Art Project, given its closed architecture – the system interacts with other Google applications, enhancing Google user profiles with the Art Project options, but does nothing to export such information to third-party applications. However, even Europeana, with its transparent, open-access policy, is not explicit about allowing users to extract their profile information onto other, perhaps competitive, applications. In addition, no developments have been recorded as to the interoperability among Europeana and the Google Books platform, something which would have offered users substantial benefits.[55] It appears therefore that, as far as a right to interoperability is concerned, users will have to wait until the online content aggregators are well-established and feel confident enough to start opening up their Internet platforms to competitors.

7. Museum digitised collections as Public Sector Information?

The access (and exploitation) of public-sector information (PSI) by third parties has gained exponentially in importance in the EU during the last decade. Being ultimately connected both to modern governance best practices and considered an instance of public sector transparency, as well as an investment opportunity, the field has witnessed both specialized legislation[56] and intensive networking. However, it is with great surprise that the New Renaissance 2011 report takes the application of the PSI Directive for granted: one of its key recommendations is that “Where cultural institutions charge private companies for the re-use of the digitised public domain material, they should comply with the rules of Directive 2003/98/EC on the re-use of public sector information”[57]. Despite the Report’s efforts to facilitate content re-use by reference to the PSI Directive, such an interpretation does not seem to be justified under the current PSI legal framework.[58] This is mainly due to an explicit exemption from the PSI Directive scope: as per its wording, “this Directive shall not apply to […] documents held by cultural establishments, such as museums, libraries, archives, orchestras, operas, ballets and theatres”[59]. In addition, “documents”, for the same Directive’s purposes, are to include “any content whatever its medium (written on paper or stored in electronic form or as a sound, visual or audiovisual recording)”,[60] thus including the digitised files of a museum collection (and, of course, any digital-born material as well).

Neither are the Directive’s Recitals pointing to any other direction: as for the “document” definition, it is stated that “a generic definition” of the term is aimed at, “covering any representation of acts, facts, or information” and any compilation therefrom, whatever the medium.[61]

It is therefore only with extreme difficulty that application of the PSI Directive on digitised museum collections could be supported. Such an interpretation of the Directive would perhaps imply that the above exemption refers only to original information and not to its subsequent recreations for the digital environment. As a result, although museum collections are excluded by themselves, their digitised reproductions are not. Again, this would be a wide-reaching reading of the PSI Directive, given the explicit exemption and the wide definition of the “document”.

Evidently, also the “public body” requirement creates difficulties as well: whatever its interpretation, the PSI Directive applies only to public bodies, and so private museums (and also material with its own third party-copyright[62]) fall, by definition, outside its scope.

Consequently, it is to be assumed that the PSI Directive, at least under its current wording, will not affect online museum content aggregation in any way (and, in any event, its purposes for enabling re-use of such information is effectively fulfilled by the Europeana Foundation or Google online content aggregation).

8. Restitution, and other financial considerations

8.1 Is online museum content aggregation financially sustainable?

Despite the heroic entrepreneurship behind online museum content aggregation, financial considerations are inevitable. Notwithstanding its public good and noteworthy purposes, the fact remains that the maintenance of complex websites and services, such as the ones discussed in this article, require substantial financial resources. Given that, at least for now, income is not derived directly either from users or from content providers, the question is, who shall pay for these websites to remain operational and to develop further?

We see that both models provide no straight-forward answers to the self-evident question of their financial sustainability. Europeana is funded by the European Commission for the time being, but it is not yet clear if it is envisaged that this will continue to be the case for as long as it exists. There are presently no mechanisms in place for Europeana to create income directly, either from users or from participating museums and other content providers.

The Google Art Project has yet to show its true implementation. It could potentially function as an additional income maker for Google, exactly in the way that Google Books enter bilateral book digitisation agreements. Or, it could charge a fee to museums to present their content onto its website. Or, it could charge users for certain uses of content within their profiles. It could even feed advertisements into user profiles and webpages, as per its current, and successful, main revenue stream. Or, it could simply constitute an add-on to generic Google user accounts, offering an indirect competitive advantage in the fierce online social media market.

It appears, therefore, that both models are still at a transitory period, mostly aiming at their establishment, both in terms of content and user conscience, while sustainability considerations are left for the future. However, once income realization is achieved, regardless of whether it is direct or indirect, perhaps a revenue-share with the content providers that ultimately made such income possible would not seem unreasonable.

8.2 Could users profit financially from online museum content aggregation projects?

The sustainability of the service providers is not the only financial consideration when it comes to online museum content aggregation. The second question pertains to users, and whether it should be made possible for them to profit from such content aggregation. In other words, should users (including enterprises and application developers, considering that both models are, in fact, platforms) be allowed to put content made available therein to commercial purposes? And, if yes, under which terms and conditions through a revenue-share scheme or without any restrictions whatsoever?

Again, no clear responses are provided to this question. For the time being both models enable, if not encourage, users to create Web 2.0-like profiles onto them, expressly for non-commercial purposes. What additional uses will be permitted in the future, when content reaches a critical mass, remains to be seen. Interoperability requests could prove crucial, particularly if users are allowed to export profiles and data onto other platforms. For now, copyright and other restrictions by the content providers continue to apply, effectively discouraging organized investment efforts: even if application (app) development is permitted, enterprises are not likely to engage unless blanket intellectual property rules are applicable on all content made available through content aggregation websites.

As far as Europeana is concerned, although the Commission in principle encourages further commercial uses of digitised cultural content[63], its terms of use expressly prohibit any “sharing” or “transformation” of its website “for commercial purposes”.[64] Surprisingly, this blanket exclusion is not fully justified by its Agreement: in effect, the metadata provided under the Agreement are actually subject to commercial uses, while thumbnail images are not.[65] Why the terms of the Agreement are not depicted exactly on the Europeana website, and why commercial uses of its content are altogether excluded despite the Europeana Foundation right to do otherwise, is unclear. In any event, should such commercial use actually be allowed, it remains to be seen whether this shall be achieved under a revenue-share scheme or not. For the time being, the Agreement does not state anything to this end, enabling thus further commercial uses free of charge, at least as far as content providers are concerned. On the other hand, it appears that museums are critical of any income-making use of data they have provided to Europeana free of charge[66].

It remains to be clarified whether the act of transferring content to Europeana by museums and other content providers should signify, at the same time, a resignation from their rights to restitution in cases of commercial uses of such content. If museums have a policy within their own websites that content is provided only for non-commercial purposes, it is hard to accept that the same content, when accessed through Europeana, suddenly loses this protection. In addition, no legal instrument in effect as of now makes such a resignation mandatory for museums, making the, quasi-imposed,[67] contractual obligation at times hard to justify.

On the other hand, one has to carefully assess, what exactly could be commercialized through the Europeana website: if such right ultimately pertains only to (part of the) museum metadata and even some thumbnail images, it could be argued that museums sacrifice relatively little for the public good.

Other, over-reaching arguments refer to whether museum artifacts and their digital files should be considered human culture upon which no commercial uses are to be allowed whatsoever.[68] They refer also to the fact that museum content digitisation projects are mostly paid by public funds and they therefore have to create a return of investment, through the increase of income, employment, etc. Or, to the fact that museum digital files are public-sector information (PSI) after all, and that they should be treated as such in the relevant regulatory framework. It seems that the policy discussion is ongoing, and any outcome is possible.

With regard to the Google Art Project, given its closed architecture, it is hard to imagine any non-Google commercial uses of its content. In particular, with regard to the high-analysis images, intellectual property rights remain with the museums that provided them. With regard to its navigational information, users are guided to request Google permission for any commercial uses of its contents. Unless, of course, Google opens up its platform to third party developers and users alike (in the growing mobile phone applications market, for example), although even then, commercial uses of the platform shall at best remain indirect.[69]

A final, financially-related consideration relates to cases of unlawful use of content available on the aggregating websites. It obviously cannot be excluded that users may not follow the rules and indeed use content available through the aggregating websites in ways that infringe their intellectual property rights and/or the relevant terms of service. It is yet unclear whether online content aggregators should assume greater responsibility while monitoring their users, and whether museums have the right to take action against them in cases of their rights infringement by their users, or whether the general, e-commerce, rule shall apply on provider liability. In any event, a relevant contractual, liability partitioning clause should be clear and straightforward, to both content providers and to Internet users, in the relevant agreements.[70]

9. Conclusion

The Internet constitutes, by now, both an opportunity and a cause for concern for museum managements around the world. While it enhances public access, it requires substantial resources in the creation and maintenance of a contemporary, visible and user-friendly website. This task, in addition to the actual digitisation process of museum collections, proves to be an ongoing, expensive and difficult project in itself. Online museum content aggregation could lift at least the Internet-related burden off of museums’ managements’ shoulders. By uploading or linking their content to the relevant websites, museum managements are reassured that their collections are treated in the best possible way in the online world, benefiting from increased visibility and performing their public good purposes. User access is enhanced. At the same time, common digitisation standards are promoted, an indispensable tool for enhancing public access that is attained only with difficulty among contemporary (mostly piecemeal) and standalone museum collections’ digitisation projects.

Consequently, online museum content aggregation constitutes a worthy cause. The eventual type and form that such Internet content aggregation shall assume still remains to be seen. Currently, two models stand out among their peers in terms of content volume and user exposure, and thus attracted the attention of this analysis: Europeana and the Google Art Project. These two models are not competitive. Certainly, Europeana was developed upon digital library premises, with specific reference to the Google Books initiative; but it was expressly “not conceived as a competitor for digitisation projects in the private sector”,[71] a role that could very well be assumed by the Google Art Project. The latter, from its part, followed the launch of Europeana but has adopted a completely different approach and solutions (having international rather than European character. However, nothing precludes that, in the future, new entries shall occur in the field, or even that these mentioned above shall change their operation models for their own reasons and purposes.

Despite the transitory stage, a point that has already             become clear is that the regulatory framework in effect today is not entirely suited to properly accommodate the new reality. At least in Europe, online museum content aggregation is affected by several fields of law. Supposing that proprietary issues over their collections have been resolved by each museum individually (an assumption that may prove rather optimistic in practice), apart from the basic question of copyright applicability on digitised photos of artifacts, the sui generis database right as well as the PSI re-use regulations (regardless whether by way of exemption or not) are all applicable at the same time. Provider liability and provisions over any commercial re-uses of data are also relevant. In view of the above, it would appear that an ad hoc regulatory framework is urgently needed; until that time, all these issues are apparently addressed in the relevant agreements between museums and content aggregators, but this lacks clarity and transparency and cannot thus constitute a long-term solution.

At a more abstract level, the underlying power shift cannot remain unnoticed, because it bears, among others, concrete legal consequences. Online content aggregation disrupts the known museum use model. Until today museums operate in the real world, making available whichever part of their collections they wish to visitors who pay an entry fee for such use. Further use of their collections (for instance, photos by visitors) is subject to strict, museum-imposed rules. Museum proprietary websites are ultimately extensions of the above, offline, access model onto the Internet. An online content aggregation model threatens to change everything, because it affects the basic notion of public access. Visitors no longer visit the museum itself, or its website, in order to gain access and use its collections. Consequently, decisions on a collections’ listing, appearance etc., that used to be the prerogative of museum managements, are now made by a newcomer in the field, the online content aggregator.

This shift of the decision-making process bears already identifiable risks for museums. Online content aggregators aim at maximizing their user visibility and establishing a concrete Internet presence. If they achieve their goals, a potential failure of a museum to list its collection under the relevant Internet platform(s) may ultimately result in that collection failing to register with public knowledge. In addition, entry into an online content aggregating platform, per se, could prove problematic for museums. Under the two models examined in this article, the relevant co-operation agreements are in one way (formally in Europeana) or another (commercially, if the Google Art Project takes off) non-negotiable. The relationship therefore risks turning from a co-operation between equal parties to a conflict between platform operators and skeptical but obligated users. Furthermore, it is unclear whether, once uploaded, content may be “suspended” (become unavailable) from the content aggregation website for whichever reason. Rules on content listing (for instance, homepage rotation or not) are, for the time being, left with the content aggregators and their own decision-making models.

It is also important to note that museum content, once aggregated online, will most likely gain an independent existence. Online aggregators, even under their best intentions, cannot possibly warrant a restricted non-profit (or, for the same purposes, profitable) use of museum collections on their websites. Users (and even application developers) are usually keen to circumvent technical measures and contractual obligations. It is therefore to be expected that, once aggregated online, museum content use shall potentially exceed the limitations and purposes, of the museums that released it.

In light of the above, it is highly likely that the ad hoc regulatory framework that will eventually be required to deal with online museum content aggregation shall have to practically protect museums’ interests from content aggregators’ business practices and related risks.

Content aggregators themselves have difficult issues to resolve. They have to boldly face a complex and perhaps unaccommodating regulatory framework that even, as is the case for the Google Art Project, transcends jurisdictions with substantially different rules. They also have to deal with their sustainability issue, a problem further complicated by the fact that they appear at times to be trying to commercialise on human culture and heritage. Despite their heroic entrepreneurship, public acceptance and strong support by the state (in the case of Europeana) or the museums themselves (in the case of the Google Art Project) these problems may ultimately bring about their demise.

Because online museum content aggregation is still at a transitory stage, whereby platform development and content digitisation and uploading are the only truly pressing issues, perhaps a careful assessment of options and future strategies is not yet on the top of either the providers’ or the museum managements’ agenda. The legal framework to-date follows such indecision. This is, however, a risky policy: Unless a user-friendly, straightforward and transparent policy is developed, backed up by the appropriate regulations, a worthy and potentially added-value initiative could be effectively undermined before it has its chance to develop its full, public good, potential.