Introduction

The use and reuse of digitized cultural heritage in research, creative, or commercial settings is often hampered by copyright: in particular, where copyright status or permissions cannot be ascertained. Orphan Works make up a significant portion of the material collections of Galleries, Libraries, Archives, and Museums (GLAMs), both in the United Kingdom and beyond. Orphan Works are those works protected by copyright and for which the rights holder is unable to be identified or, even if identified, cannot be located. These works encompass all different types of material culture: artwork, still images, sound and film recordings, as well as published and unpublished texts. Previous studies have shown that the mission of the cultural heritage sectors to provide access to this vast array of materials is stymied by a lack of clear copyright legislation, which is exacerbated in the digital domain. On 25 October 2012, Directive 2012/28/EU was passed by the European Parliament and the Council, with specific provisions for Orphan Works (European Parliament and the Council, 2012). The purpose of this article is to trace the enactment of EU Directive 2012/28/EU within the United Kingdom via the implementation of the UK Orphan Works Licensing Scheme (OWLS) from October 2014, and to ascertain how the OWLS has been received by the UK GLAM sector. This research addresses two questions: one, how is current EU Orphan Works legislation affecting the cultural heritage sector in the United Kingdom including the use of digitized content; and two, what changes can be made to the implementation of the EU Directive in the UK to better support the mission of cultural heritage institutions, including serving the research and creative communities?

To address these research questions, we discuss key issues in the Orphan Works problem by summarizing previous UK studies that contributed to evidence-based recommendations aiding in the creation of the EU Directive and the subsequent UK Orphan Works Licensing Scheme. We provide a brief analysis of the EU Directive as compared the UK Orphan Works Licensing Scheme, and conclude with suggestions for improving the Scheme based on feedback from our qualitative study conducted between December 2015 and February 2016, and our own experience of using the Scheme.

Orphan Works and the Cultural Heritage Sector

Cultural heritage institutions are faced with a growing tension: they are increasingly expected to provide access to their collections in an online environment that encourages reuse, remixing, and the open licensing of content, which should provide users with the permissions to do so (Terras, 2015). However, these institutions have a wealth of Orphan Works – materials where the copyright owner is not known or cannot be traced (Simone, 2014) that they cannot disseminate because by doing so, they will violate copyright law. UK copyright specialist Naomi Korn found that 20% of the British Film Institute’s holdings constitute Orphan Works, while the Imperial War Museum holds approximately two million archival photographs and artworks that are Orphans (Korn, 2009). The National Archives, the National Records of Scotland, and the British Library have estimated that 40% of their total collections are Orphan Works (Rosati, 2013). While these collections may be vast, the unknown variable of potential legal action means that memory institutions funded by the public sector have traditionally had a low-risk appetite for opposing government legislation.

Online dissemination of digitized collections is increasingly becoming an expectation rather than a luxury at GLAM institutions (Hughes, 2003; Rikowksi, 2011). As digitization and use of online collections has grown, so has a research interest in how digitized collections are being used by online audiences (Hughes, 2011; Gooding, 2016). As a result, staff are beginning to challenge the status quo of how institutional and legislative barriers surrounding Orphan Works are cutting off large swathes of holdings from being disseminated to a potential world-wise user-base. In response to the implementation of the EU Directive in the UK at the end of October 2014, the National Library of Scotland and many other memory institutions in the UK launched the ‘Free Our History’ copyright campaign, displaying exhibits of empty cases and blank sheets of paper in lieu of 20th-century materials, stating that they would love to share the materials with the public, but were unable to due to copyright laws (HistoryScotland, 2014). The rationale behind this protest and the larger tension in the sector is that by following copyright legislation and by stemming the flow of the digitization of Orphan Works, memory institutions and their staff come into direct conflict with their missions to support and facilitate the dissemination and reuse of cultural heritage materials. Such goals are explicitly stated by the mission statements of the British Library (British Library, 2010), the National Library of Scotland (2015), the National Library of Wales (2010), and myriad other GLAMs across the UK. Cultural heritage institutions face the demand for digitization and online dissemination of works of educational, historical, or cultural value at a relatively low cost. While orphaned sound and film recordings can be preserved and copied to new formats as a conservational method, these materials still cannot be legally accessed or disseminated to the public due to copyright restrictions (Commission of the European Communities, 2008). The cultural and research content held in European memory institutions has been estimated to have a market value of EUR 27 billion, representing ‘the biggest single information content resource for the creation of value-added information content and services’ (Jančič et al. 2014: 353) but the majority of 20th-century content, much of which is Orphan Works, cannot be utilized by either institutions or a wider audience.

Although there has been some recent institutional engagement with repackaging digital GLAM content beyond traditional institutional catalogues – such as the Design Collection at the National Archives (2018) which curates ‘rare beautiful design to inspire your own creations’ as part of a subscription service, or the work of British Library Labs (2018) to encourage and formally recognize creative use of their digitized content – large-scale user uptake and reuse of digitized material has been low. It has been suggested that this is partly due to the confusion over rights and licensing that accompanies much digital content, particularly those materials which are Orphan Works. As a result, ‘memory institutions have not been able to fully adopt digital technology in order to become part of the information economy… to date, there is little known about the extent to which heritage organizations are able to innovate’ (Borowiecki and Navarette, 2016: 227–8). By clarifying legal frameworks and institutional positions for reuse of such digital content, the GLAM sector would be supporting traditional research access to their collections while simultaneously supporting new and innovative uses, including design creation, remixing, and monetizing collections. While moral and legal copyright must be respected, the barrier which the Orphan Works issue has put in place and which is perceived as insurmountable, is one which must be addressed by the government and by the sector. The problem of Orphan Works is a significant and long-standing issue for collection building and dissemination in memory institutions and needs to be redressed, as institutions cannot reach out to potential user communities and encourage adoption and reuse of modern or otherwise in-copyright material, barring institutions from engaging fully with the research community, discouraging the use of collections in research outputs, and hampering future engagement with the commercial or creative industries.

The Orphan Work Status

While Orphan Works present a complex set of problems for staff of cultural heritage institutions, copyright specialists, and intellectual property officials across Europe, the origin of the problems stems from a lack of documentation about an individual work or collection. The reasons behind this missing information can be varied, intricate, and are often an unintentional consequence of natural life and business events. For example, publishers or other manufacturers may go into receivership, leaving no contacts for content generated from authors, artists, or illustrators; the estate of creators may not be identifiable years after their death; illustrations, designs, and artwork may not have any identifiable information leaving the original creator unknown (and often unknowable); et cetera. Many of these situations result in individuals being unaware of their status as copyright holders of a work. When records of ownership are lost, chains of provenance are broken, and when items deposited in memory institutions have no metadata regarding ownership, provenance, or copyright, the creation of Orphan Works is a natural effect.

It would be unreasonable to dictate that all memory institutions should focus only on the digitization or display of items where copyright was unambiguous, but the significant barriers surrounding the processing and display of Orphan Works has effectively created that situation. This is certainly the case in some memory institutions given small budgets and a desire to avoid any possible liability. However, there is a disconnect between the imperative placed on GLAMs to participate in collection building and safeguarding community memory on the one hand (contributing digital assets to the Open Glam movement),1 and the government’s role of establishing copyright licenses to allow further dissemination of these objects, texts, and artifacts via digitization on the other.

Literature Review

To better understand the response of the cultural heritage sector to current UK and EU Orphan Works legislation, we performed a small survey of stakeholders. In preparation for the design of the survey, we conducted a literature review of previous studies leading up to the EU Directive and the resulting UK Orphan Works Licensing Scheme (OWLS). The reports showed the complexity of the issues and played a key role in helping us to pinpoint issues in the Orphan Works debate that we could address in our survey.

Empirical Studies on the Orphan Works Issue Prior to the 2012 EU Directive

Since the mid 2000s, the extent, importance, and complexity of the Orphan Works problem within the UK copyright context has been explored in quantitative and qualitative reports including the Gowers Review of Intellectual Property (2006), the Joint Information Systems Committee (Jisc)/Naomi Korn ‘In from the Cold’ report (2009), the British Library/ARROW ‘Seeking New Landscapes’ report (2011), the Digital Opportunities/Hargreaves report (2011), and the UK Intellectual Property Office Orphan Works Impact Assessment report (2012). These provide an outline of how responses from the cultural heritage sector led to the development of the EU Directive in 2012.

The Gowers Review (2006), sanctioned by HM Treasury, called for evidence to form the basis of a report to provide a UK perspective on the wider context of EU intellectual property law. With 517 responses, the review came up with several Orphan Works-related recommendations for the UK Intellectual Property Office (IPO), broadly expressing an urgent need to update intellectual property standards to fulfill the public mission of cultural heritage institutions and for intellectual property legislators to consider the ease with which the digital medium could bring together material cultural collections and wide-ranging groups of users. The Orphan Works-specific recommendations included ‘clear guidance on the parameters of a reasonable search for Orphan Works’, as well as the establishment of a ‘voluntary register of copyright, either on its own or through partnerships with database holders, by 2008’ (HM Treasury, 2006). The report also noted that although the British Library had suggested it be up to individual institutions to determine what would constitute a diligent search, these institutions (and users for that matter) ‘may still need a guide as to what constitutes a thorough search to protect them from liability’ (HM Treasury, 2006: 71). The British Library’s response to the review asserted that ‘intellectual property must not hinder research and innovation’ and that ‘there would be a tangible economic and public benefit if a provision were established to streamline the process of seeking rights clearance to deal with the use of Orphan Works whereas at present many works are, arguably, unnecessarily “locked up”’ (British Library, 2006: 5). This response was echoed by the submission from the Research Information Network (RIN), a ‘consortium of the four UK higher education Funding Bodies, the three UK National Libraries, and the eight Research Councils’ (2006: 1). The RIN argues that there were ‘often insurmountable difficulties’ to tracing rights-owners, presenting ‘a major barrier to the kinds of digiti[s]ation projects that would transform access to research collections of fundamental importance to social, historical, and cultural studies’ (2006: 7). The British Broadcasting Corporation (BBC) further legitimized this claim by contending that their staff experienced ‘significant difficulties when trying to identify and/or trace owners of copyright’ with its own archival content (2006: 7). A response offered by the Arts and Humanities Research Council (AHRC) Research Centre for Studies in Intellectual Property at the University of Edinburgh reasoned that confusing copyright law ‘both threatens to and does create impediments’ in the development and dissemination of digitization projects and that there is uncertainty in the research community about what constitutes ‘non-commercial research’ and what exactly are ‘the boundaries of fair dealing’ (University of Edinburgh, 2006: 2). The authors reasoned that in the academic sphere, the ability to move from ‘criticism and review’ to ‘analysis’ is a necessary one, and that there is real concern about whether research garnering some modicum of royalties, but which is designed for the pursuit of academic inquiry, counts as ‘non-commercial use of Orphan Works’ (University of Edinburgh, 2006: 2).

In 2009, Jisc published the results of a mass survey conducted in the United Kingdom by copyright specialist Naomi Korn (2009), which uncovered the extent to which problems related to Orphan Works were affecting the cultural heritage sector. Responses from 503 members of the public and private sectors revealed that the problem was much larger than initially anticipated, and that staff awareness about the extent of the problem was ‘often limited’ (Korn, 2009: 7). Korn found that Orphan Works in the UK range from items of potentially high commercial value to those of low commercial value but ‘high academic, cultural and historic work, such as documentary photographs, letters and sound recordings, where a recognized rights holder is unlikely’ (Korn, 2009: 5). In some cases, the vast majority of materials in a collection were of commercial value but the items of the highest perceived value to scholarship were Orphan Works, and therefore untouchable. Digitization of these materials would be costly, with rights clearance being an additional cost that would make the project infeasible. As one respondent noted:

Orphan Works are currently automatically excluded from any digitization project conducted within the organization, owing to the complexity and time required to trace ownership. This has a marked effect on the accessibility and scope of collections (photographic, archival, and sound recordings in particular) made available to the public (Korn, 2009: 56).

Respondents to the Jisc survey also reported either entirely avoiding Orphan Works in their digital collection(s), or, in far fewer cases, adopting a risk management approach to displaying the works. This risk-management approach included displaying works with a take-down policy if a rights-holder claimed their rights and asked for an item to be removed.

This combination of problems was amplified by the lack of clarity regarding what exactly constitutes a diligent search to locate copyright holders. It is unclear how much and what type of documentation should be kept in order to prove a diligent search has been made, and what legal certainty such documents hold against potential indemnity for cultural heritage institutions. Respondents reported having Orphan Works in their collections that included but were not limited to books, theses, unpublished manuscripts, photographs taken in the UK and abroad by British photographers, materials received from abroad where copyright ownership could not be established as either British or foreign, materials created during wartime where records pertaining to copyright were destroyed, architectural drawings, and feminist collective literature where only one copyright owner of the collective could be traced, whereas all members’ consent would be needed before digitization and dissemination could go forward (Korn, 2009: 51–2). Survey respondents reported that materials for which rights of ownership were relatively easiest to trace were fine art works, as they generally carried a chain of provenance leading back to creation (Korn, 2009: 13).

Complex copyright issues, lack of clear guidance on due diligence procedures, and risk-averse cultural heritage institutions have led to what Korn deemed a ‘black hole of 20th- and 21st-century content’ (2009: 24). She concluded that without significant legal intervention, this ‘black hole’ would continue to plague cultural heritage institutions and their mission of digital dissemination. Korn further recommended the development of a licensing scheme and a national database that cultural heritage organizations could opt into if they saw fit. The report recommended a risk-management approach and indicated a need for comprehensive training opportunities for staff, including straightforward workflows and best practices for due diligence, as well as better understanding of how copyright law affects the public sector (Korn, 2009: 28). Some survey respondents requested the removal of Orphan Works from the UK Copyright Act of 1979, which would grant cultural heritage organizations full amnesty to use these works and create digital facsimiles available for non-commercial and educational purposes. This would be in keeping with the memory institutions’ missions and would, Korn argued, be welcomed by rights holders who did not realize that they owned the copyright of said works (2009: 29).

In 2011, as part of the EU-funded ARROW (Accessible Registries of Rights Information and Orphan Works towards Europeana) project,2 the British Library commissioned Barbara Stratton to conduct the ‘Seeking New Landscapes’ report which explores rights clearance issues for books in the British Library’s collections. The ARROW project included organizations and specialists aiming to create a database of rights information for printed books of all types: in-copyright, out-of-copyright, out-of-print, and orphaned. Stratton examined the process of due diligence for rights clearance on a random selection of 140 books in the British Library’s holdings published between 1870 and 2010, with ten books from each decade (2011: 4). Of the 140 books in the sample, 43 were Orphan Works, which equated to 31% of the total sample. Stratton concluded that ‘on average it took four hours per book to undertake “diligent search.” This involved clarifying copyright status of the work and then identifying rights holders and requesting permissions’ (2011: 5). These results indicate that a manual diligent search process for individual books in a collection the size of the British Library’s is infeasible. Taking into account that the sample included only books, and the UK GLAM sector’s larger institutions also include Orphan Works across the spectrum of material culture, Stratton’s research supports the conclusion that a diligent search system is not conducive to efficiently determining the copyright status, contacting rights holders, and requesting permissions for digitization and dissemination. As a result, dissemination is stymied, research is stultified, and collections cannot benefit from the critical work of cross-contextualization, or newer opportunities such as Linked Open Data across institutions and collection, never mind taking advantage of more creative reuse opportunities.

The ‘Seeking New Landscapes’ report compared Stratton’s manual rights clearance process with the use of the ARROW system, which is intended to do much of the legwork of diligent search by running available printed books metadata through publishers’, institutions’, and political organizations’ databases for a fee (Europeana, 2014: n.p.). However, as the system is designed to support diligent search exclusively for published books, it does not address the process of rights clearance for different materials, nor are such databased helpful for unpublished texts, or still and moving art works. Indeed, the British Library previously noted in 2006 that ‘published books are relatively easy to seek permissions for – in comparison to audio recording or film with their multiple layers of rights, or photographs which generally have no information attached to them’ (British Library, 2006: 25). The Jisc report that preceded the Gowers Review determined that efficient rights clearance mechanisms are needed for cultural heritage institutions to fulfill their public mission of dissemination and access, and for those institutions to navigate due diligence requirements in relatively quick and legal ways.

The Digital Opportunities Report, more commonly known as the Hargreaves report, was prepared by Professor Ian Hargreaves of Cardiff University, UK in 2011. Hargreaves used economic trends and case study evidence as empirical data to argue that ‘the problem of Orphan Works…represents the starkest failure of the copyright framework to adapt to the digital landscape’ (2011: 38). Indeed, Hargreaves concluded that copyright law is not able to keep up with the demand for digital dissemination of works, arguing that ‘opening up Orphan Works is a move to which there is no national economic downside’ (2011: 39). Much like the Jisc report, the Hargreaves report concluded that the Orphan Works issue severely impedes mass digitization efforts in the GLAM sector. Hargreaves ultimately recommended that ‘[t]he [UK] Government should legislate to enable licensing of Orphan Works. This should establish extended collective licensing for mass licensing of Orphan Works, and a clearance procedure for use of individual works’ (2011: 40).

Taking into account these previous studies, the UK Intellectual Property Office published the ‘Orphan Works Impact Assessment’ in June 2012. This report aimed to determine the cost and rationale for setting up and authorizing a body for commercial and non-commercial licensing (at different rates) of Orphan Works, and ultimately recommended a service wherein Orphan Works, for which a diligent search had been approved, could be listed and publicly accessed. The report points to interviews with members of the cultural heritage sector as evidence that GLAMs are disproportionately affected by the Orphan Works problem and that institutional holdings largely consist of unpublished documents (UK IPO, 2012). In the majority of instances when these holdings have been displayed or made available to the public, ‘rights-holders do not reappear, and those who do have been pleased to see the work brought to public attention’ (UK IPO, 2012: 2).

The impact assessment suggested that a license rate, lasting for a maximum of five years, should be a fixed fee and ‘proportional to the intended use’ of the work (UK IPO, 2012: 6). Ironically, the report went on to argue that ‘a scheme which required individual authorization via a bureaucratic procedure is likely to be very little used’ (UK IPO, 2012: 5). A potential plan for spot-testing rigorous due diligence was mooted (UK IPO, 2012: 5), but it did not take into account the fact that more information about rights holders could appear over time. This is a fundamental problem with the idea of diligent search, and a driving rationale for eschewing licensing systems as they currently stand now: Orphan Works should be made available with a rigorous take-down policy instead, so that if, over a period of years, more information about the Orphan Work becomes available, a rights holder can contact an institution and ask that the work be either credited or removed.

A major issue with the impact assessment is that it does not adequately address the issue of disseminating works and building digital collections in a timely manner. For example, the report gathered ‘probably the most complete list of Orphan Works estimates that ha[d] been collected [up to 2012]’ (UK IPO, 2012: 9) and found that sample holdings in the UK encompassed roughly 172 million works, 27% of which constitute Orphans. The estimate and the time required to clear rights across different types of media, through different collecting societies with varying degrees of accuracy and documentation, organizational, and institutional databases, means that in order to conduct a diligent search would collectively require many millions of hours. The estimate cost of these diligent searches for the BBC and the British Library, two of the largest collections in the UK, range between £6.1 and £7.3 million (UK IPO, 2012: 9). The impact assessment estimated that given the resources required, only 5–10% of the British Library’s and BBC’s Orphan Work holdings would feasibly be digitally disseminated in the near future.

Further, the impact assessment asserts that the ‘Orphan Works system we propose is one which would cover only use in the UK, as we cannot permit use outside the UK’ (UK IPO, 2012: 9). This recommendation, implemented in the resulting UK Orphan Works Licensing Scheme released in October 2014, directly contravenes the idea of the free movement of people and goods in the EU, and certainly does not account for the ease with which digital medium can be distributed, regardless of territory, across the Internet. It is a prime example of how copyright law has not been able to keep pace with digital innovation, and why collections in cultural heritage institutions, and by extension researchers, the public, and also the creative industries, are suffering as a result.

Implementation of the EU Directive 2012/28/EU with Provisions for Orphan Works

Reports about the Orphan Works problem prior to 2012, along with those submitted by other EU Member States, laid the foundation for what would become EU Directive 2012/28/EU with Provisions for Orphan Works. The Directive was established to lay out the mutually agreed legitimate uses and rights clearance processes of the Member States for both commercial and non-commercial uses of Orphan Works, so that one European system could be established to unify the dissemination of works digitally, and so that it could be attractive to businesses like Google in its mass digitization of printed books (Rosati, 2013). Setting out common rules on the digitization and online display of Orphan Works, the EU Directive offers guidance on due diligence, exceptions, permitted use, and the generation of revenue by institutions using Orphan Works from their collections (European Parliament and the Council of the European Union, 2012; IPKat.blogspot.co.uk, 2012). However, the final text of the Directive did not include provisions for commercial uses of Orphan Works, stating instead that the Directive was intended for the benefit of public educational organizations like GLAMs. Further, the major flaw with the Directive is that it does not provide an adequate explanation of what appropriate due diligence entails; so that each Member State must create these requirements on their own (IPKat.blogspot.co.uk, 2014).

Immediate reactions to the EU Directive and its potential implementation in the UK were not positive. The ‘1709 Blog’, which focuses on issues of copyright across the UK and EU, was particularly critical, publishing a series of posts about the implications of the Directive across Europe, and how it might complicate matters regarding Orphan Works rather than solve them, specifically because the Directive left more complete legislation up to each Member State (1709Blog, 2014a, b, c). Wikimedia also responded negatively, saying that the Scheme and the Directive brought about ‘nothing new’ and that more radical change was needed (Wikimedia Blog, 2014). TechDirt – a blog about the intersection between government policy, copyright, and innovation – was more direct, publishing an article just a week after the Scheme went into effect, entitled ‘UK Launches Orphan Works Licensing Scheme, Misses Huge Opportunity to Make It Much Better’ (TechDirt, 2014). The article quoted the Chartered Institute of Library and Information Professionals (CILIP) and cited the fact that the UK government did not fight to remove the 2039 provision, which inhibits cultural heritage institutions from disseminating unpublished works until that year.

The EU Directive was the impetus for the creation of the EU Orphan Works Database.3 The EU Orphan Works Database is open only to cultural heritage organizations and only available to record the non-commercial use of Orphan Works (see Figure 1 below for a detailed breakdown of the differences between the EU Directive and the UK Orphan Works Licensing Scheme). To record an Orphan Work in the EU Orphan Works Database, an institution must first register. As of October 2018, the Database includes over 6,000 works across a range of media, registered by institutions across Europe, including the British Library, the British Film Institute, and the UK National Archives (European Union Intellectual Property Office, 2014). In contrast to the EU Orphan Works Database, the UK Orphan Works Licensing Scheme allows for both individuals and organizations to register, and also allows for the non-commercial and commercial licensing or Orphan Works, at a license fee set by the UK Intellectual Property Office, and based on the way that the work will be used (commercial vs. non-commercial). The major difference in terms of works that the EU Directive covered was photographs, due to considerable opposition from creators across the European Union about this provision. In this case, the UK Orphan Works Licensing Scheme allowed for more leeway in terms of work types, but angered the photography community (who are worried about images being stripped of metadata and licensed for redistribution) as a result (BBC, 2013). This has since been revisited, and the EU Orphan Works Database now includes photographs.

Figure 1 

A breakdown of the differences between the UK Orphan Works Licensing Scheme, here labeled ‘IPO License Scheme’ on the left, and the EU Directive on the right. Reprinted from O’Connell (2015: 49).

The reports outlined above provided us with an understanding of the main issues facing cultural heritage institutions with regard to Orphan Works. First and foremost, the issue that we determined is the key concern for the sector is diligent search, otherwise called due diligence. While the concept of diligent search was analyzed in many of the preceding reports, its effect on the GLAM sector is multi-layered and complex.

Current Approaches: Diligent Search

The largest roadblock to the dissemination of Orphan Works is the time, cost, and complexity of diligent search (Vuopala, 2010; Ringnalda, 2011; Hansen, Hinze & Urban, 2013; Schroff, Favale & Bertoni, 2017). Diligent search is the process of looking for rights holders through various information sources and available data about previous rights holders and chains of provenance (Van Gompel, 2007; De la Durantaye, 2011). This process can prove more difficult in the case of audiovisual materials which often have several ‘creators’, and in the case of correspondence and other items where copyright was never renewed or asserted because the items in question were never intended to be made public. However, such items can hold high intrinsic value as the basis of academic scholarship, and where this is the case, Orphan Works should be made available through the cultural heritage institutions that hold them (Korn, 2009). Such access could profoundly shift the makeup and size of digital collections, and lead to unforeseen avenues of research.

When Orphan Works are without known rights holders and the due diligence required to find potential rights holders is both burdensome and costly, these works languish without the benefits of study or reuse. While the UK law for published works is that copyright expires 70 years after the death of the author/creator, the UK Copyright, Designs and Patents Act of 1988 adds an extra layer of complexity for unpublished works. Under a provision in this act, unpublished works benefit from copyright until 2039, meaning that unpublished works cannot be disseminated or reused digitally because doing so would violate a retroactive copyright applied generations after the work was created (UK IPO, 2015a). Across the UK GLAM sector there are millions of works covered by this protection, spanning centuries. The Government made the decision to uphold this provision in October 2014, at the same time as the implementation of the UK Orphan Works Licensing Scheme, further reinforcing copyright burdens on institutions, rather than alleviating them. The Scheme, along with the EU Directive, was marketed as a way around the 2039 law (UK IPO, 2015a). However, the larger point was lost: if the 2039 provision had been removed, the institutions would not have had to potentially dedicate millions of pounds sterling in time and staff resources to manually enter works they effectively already owned into a database in order to pay license and application fees to disseminate said works to the public. The flaw in this logic is apparent: the public is funding cultural heritage institutions to pay the Government to check the institutions’ due diligence on searching for potential rights holders for materials that are (in some cases) hundreds of years old, so that the Government can then issue a temporary license to the institutions who can, in turn, display the works to the public, who affectively already own the works. This situation benefits neither the rights holders, nor the potential users of these works, who cannot access a significant portion of cultural heritage materials due to lacking the resources needed in order to go through this process. Perhaps most significantly, this arrangement does not benefit the institutions entrusted with safekeeping Orphan Works, as the institutions cannot use them for digital promotion, as research support, for public engagement, or (if the institution elected to use the commercial options of the UK Orphan Works Licensing Scheme) for income generation. Because cultural heritage organizations have neither the time, the staff, nor the money required to perform due diligence on each individual Orphan Work in a collection of – in some cases – many millions of items, the immediate cost cannot be overlooked when weighing the eventual benefits of such an enterprise. The EU Directive and the resulting UK Orphan Works Licensing Scheme were meant to address and alleviate these burdens, essentially transforming the cultural heritage landscape so that millions of works could be digitized and made available in a timely manner.

Pre-Survey Study: A Use Case on Micro-Scale Commercial Licensing of an Orphan Work

Aside from the literature review we conducted about the previous reports, one of the authors of this article also performed a use case of the UK Orphan Works Licensing Scheme for commercial licensing. While the use case and the application process is documented in detail elsewhere (Terras, 2015), we provide a brief summary here.

One of the benefits of the UK OWLS over the EU Directive and the associated EU Orphan Works Database is that the OWLS allows for commercial licensing of Orphan Works. Just after the OWLS was announced in October 2014, Terras began the process of applying for a commercial license for a 1960s lantern slide entitled It’s Lolly Time!, held in the collection of the National Library of Scotland (see Figure 2 below). The intended purpose of the license was to digitize the slide and create a repeating pattern of the slide on some fabric. This fabric would be made into a scarf and/or as a roll of fabric to be sold on the online marketplaces Etsy or Spoonflower, respectively (see Figure 6 below). A diligent search for a rights holder had already been performed, but the UK IPO asked for a certification of this through a diligent search checklist. The checklist options were for the types of works that the UK OWLS currently supports: film music and sound; literary works; or still visual art. Terras selected still visual art and proceeded to fill in the checklist. At the time she began the process (2014), an impressive collection of over 50 options for databases, societies, and other avenues for search were listed. In 2018, that number had grown to 77 options (UK IPO, 2015b). Certainly, this increase must have taken considerable effort on the part of the UK IPO to put together, and it’s very useful to have the list – though providing links to each of the databases would be a helpful improvement. In the checklist form, a box next to each option requires the user to provide a narrative for how and when the option was checked, whether there was any information, or whether the option was even relevant to the Orphan Work in question (i.e., the Lolly Time slide would not be relevant to the Society of Wedding and Portrait Photographers).

Figure 2 

It’s Lolly Time!. An example of an Orphan Work. A mid-1960s lantern slide, advertising the interval refreshment at a cinema. Produced by Morgan’s Slides Ltd (which is no longer trading), illustrator unknown. Used at the Odeon Cinema, Eglinton Toll, Glasgow, although the Odeon’s records for this sort of material do not go back to the mid-60s, and in 2014, they gave permission to Melissa Terras to use the image until such time as a copyright holder comes forward. This example clearly demonstrates how difficult it is to trace permission and potential rights holders of modern in-copyright works held in institutions, and the various difference agencies that may have a stake in the process. Item held in the collection of the Moving Image Archive, National Library of Scotland. Image used with permission.

Such information helps to protect both the user and the UK IPO from claims that the diligent search was not rigorous enough. But the checklist still inevitably leaves the question open of what counts as an appropriate length of time for a diligent search. Should these databases be checked multiple times over several months? Or does a cursory glance suffice for the user to determine if the database is within scope of the potential Orphan Work? For a cultural heritage institution with multiple Orphan Works, it is clear that this process would be prohibitively expensive in time and money.

Unfortunately, it was not possible to elect to reproduce the lantern slide at an individual item level rather than a larger run of items in the commercial license application. At the time of the application (2015), the smallest run of items that a user could apply for was 5,000. Since that time, the smallest item run has been reduced to 500 items (UK IPO, 2018) as a direct result of Terras publicly challenging the higher number (S. Williams, personal communications, 19 August 2015). However, for a small-scale business or an individual user, licensing for 500 items would still be a considerable expense – not everyone who sells things on craft-led internet marketplaces has the ability to produce on that scale, and it is to the UK IPO’s benefit to support even the smallest of businesses with this Scheme. Aside from the amount of intended reproductions of the work, the OWLS application also asks users to denote what kind of work it is from a drop-down list, and a further item list, as displayed in Figure 3 below.

Figure 3 

On the left, types of Orphan Works, and on the right, subtypes of still visual art (UK IPO, 2018a).

The user selected ‘still visual art’ and then ‘illustration.’ At the time of the application, the option for ‘craftwork: image for knitting or sewing patterns’ was not offered (Terras, 2015). After selecting the options, the user indicated that the lantern slide would be used for commercial use and in particular, retailing and merchandising for apparel. In 2015 the option to use Orphan Works ‘on textiles’ had not been added (Terras, 2015), but is now an option (UK IPO, 2018). The user is then asked to indicate the size of the reproduction of the Orphan Work, from 1/16th of a page up to a full page. As Terras (2015) suggested, this does not make sense for textiles, where patterns can repeat. Re-selecting the intended use of the Orphan Work as ‘on textiles’ rather than ‘on apparel’ does not change this sizing framework, nor does changing the original subcategory of the work from ‘illustration’ to ‘craftwork: image for knitting or sewing patterns’.

After inputting all of the information about the Orphan Work, the user was asked to indicate the length of the license requested, with options ranging from three months to seven years. Selecting seven years for 5,000 reproductions and a full ‘page’ of reproduction for apparel resulted in a total license cost of £2,659.14 excluding VAT (value added tax) (Terras, 2015). Inputting the same values in 2018 yields a lower cost of £2,254.45 excluding VAT, as shown in Figure 4 below.

Figure 4 

The cost of licensing It’s Lolly Time! is lower in 2018 than in 2014 (Terras, 2015). But why?

Obviously, this cost would be outside the realm of possibility for an individual user who wanted to turn a profit on a single item or small run of items. The user opted for a non-commercial license instead, which only cost 10 pence. The £20 application fee, however, remained the same. The application was made on 31 March 2015 and the license was approved on 9 April 2015 (see Figure 5). When the UK IPO was contacted about pricing models for commercial vs non-commercial works, they replied with ‘[o]ur pricing structure runs to literally thousands of possibilities for licenses, so it is not possible to send you all costs for commercial content – it really depends on the type of orphan work and the use you want to put it to’ (UK IPO, quoted in Terras, 2015). However, this does not explain why the cost of licensing an Orphan Work would change over time, which penalizes earlier users of the Scheme, nor does it explain how the original model was chosen. An algorithm would be needed to immediately calculate the cost of a license, and the UK IPO would not supply that information (Terras, 2015). The UK IPO did note that they had used Getty Images’ pricing model for images ‘because they have publicly available information’ (quoted in Terras, 2015) but this brings up a particular question about cultural heritage organizations that are using the site to license images in their collections: is the UK IPO using Getty’s commercial image licensing structure as a factor in charging UK cultural heritage institutions for non-commercial uses? It’s an important question, and one that needs to be revisited by the UK IPO and the GLAM sector. Neither the cost calculation, nor the organizations (and previous pricing models that were consulted to create the algorithm), are supplied anywhere on the GOV.UK website. It is therefore impossible (without manually entering in all possible combinations of license types) to gauge how the UK IPO are structuring their costs, why costs are changing over time, and whether this is creating undue financial burden across the spectrum of commercial and non-commercial users.

Figure 5 

It’s Lolly Time! listed in the UK Orphan Works Licensing Scheme Register, after successfully receiving a non-commercial license. The cost of the license was 10 pence, plus a £20 application processing fee. Note the ‘Beta’ description in the top orange bar. Available at https://www.orphanworkslicensing.service.gov.uk/view-register/details?owlsNumber=OWLS000024-1&searchQuery=terras&filter=All.

Figure 6 

It’s Lolly Time! silk scarf, created by tiling the image and printing on demand via a digital printing service. The image of the Orphan Work has been licensed for non-commercial use via the OWLS which allows creation and free distribution of these materials. However, at the time of licensing, a commercial license to allow production and selling of apparel was only available for a minimum of 5,000 or less units, at a cost of £2,659.14, making the licensing of this image commercially unviable for an individual or small-scale company.

Investigating the Orphan Works Licensing Scheme: Our Approach

From a review of the previous reports we concluded that prior to the implementation of the Directive and the resulting UK Scheme, the main Orphan Works issues facing the GLAM sector were 1) confusion about what constitutes a diligent search; 2) varied understanding of copyright law among staff at memory institutions; and 3) lack of resources to carry out dissemination of Orphan Works. As these issues came up repeatedly over years of reports, we determined that gathering feedback from those institutions and individuals who had used the UK Orphan Works Licensing Scheme could provide insight into whether these concerns had been alleviated. We therefore designed and conducted a small-scale qualitative survey between December 2015 and February 2016.

Scope of our Survey

Our target audience for the survey included staff at cultural heritage institutions, intellectual property experts, scholars, researchers, and private industry professionals who would either be clearing rights for non-commercial use of Orphan Works for educational resources or for income generation through commercial licenses. We presupposed that the users of the Scheme may have varying familiarity with the UK and EU provisions, and that some may be using the Scheme on a one-time only basis to license GLAM content for reuse. Before designing the survey, our literature review suggested that institutions and individuals may be unclear about whether they are affected by Orphan Works legislation or not: most of the literature has been written by copyright specialists or legal professionals, and can therefore be opaque given the multiple coalescing legal frameworks and directives one must be familiar with in order to determine if one is affected by the issue or liable to apply for a license. Because the questions in our survey revolve around international copyright legislation, it was made available to respondents regardless of their country of origin. We reasoned that a work may potentially be housed in a country outside the legal domain of the UK, but still fall within the legal terms of the EU Directive and the UK Orphan Works Licensing Scheme, subject to rights clearance within these frameworks if the creator/author/rights holder of the work was/is British. Since many Orphan Works do not carry any information about their creators, authors, and/or rights holders, this original information about the provenance of a work means that determining the affected population for our survey (and the Scheme) was next to impossible. Thus, the issue is both UK-specific (given the terms of the license) and internationally relevant (given the dispersal of UK archival materials across Europe and around the globe). Ethical clearance for carrying out this work was given via the University College London Department of Information Studies processes: the nature of the survey work meant that there was little risk to individuals taking part.

At the time the survey was designed in October 2015, twelve months after the Scheme was launched, only 35 separate researchers, companies, and institutions had used the OWLS to register and apply for licenses of Orphan Works. The purpose of the survey was to investigate two issues. First, why cultural heritage institutions and researchers were using the Scheme, and second, why they were not. When the survey was released, these 35 people and institutions covering five countries (Canada, Germany, South Africa, the United States, and the UK) had collectively made 47 applications for 292 Orphan Works. When the survey was released, of these 292 potential works, 243 had been cleared, 32 applications for individual items were withdrawn from consideration, and 18 individual items had been received, but no decision had yet been made. Of the individual items, 46 were written works, 229 were still images, one was a film, two were musical notations, and 14 were sound recordings. None of the applications had been denied. Of the 292 works, commercial licensing was requested for 41, and non-commercial licensing for the other 252 items. Although the licensed content appears publicly listed on the UK Orphan Works Register,4 no follow-up information is provided about the exact purpose for which an Orphan Work is used after a license had been granted (UK IPO, 2014).

Dissemination of our Survey

SurveyMonkey was used to design, disseminate, and analyze the survey, which was open from 14 December 2015 to 15 February 2016. We advertised it via announcements on our Twitter handles, @merisamartinez and @melissaterras, as well as on the Humanist Listserv, the Digital Humanities Summer Institute Listserv, the Jisc British Association of Modernist Studies Listserv, and the Cultural Heritage Intellectual Property Listserv. Where public-facing contact information was available through an internet search, the survey was also sent out via email directly to those applicants who had used the UK Orphan Works Licensing Scheme. Through this broad range of dissemination circles, we were able to reach academics and scholars working in traditional departments, and various areas of the GLAM sector, and Twitter gave us a wider audience of non-academics who may have been interested in copyright issues.

Design of the Survey

The survey consisted of 33 questions and 11 pages, with ‘piping’ (automatically guiding respondents to different questions based on previous answers) between two main sections: whether a respondent had or had not heard of the UK Orphan Works Licensing Scheme. We used multiple choice questions as well as open-text response questions. The multiple-choice questions also had ‘other’ categories, and did, where appropriate, allow for multiple answer choices. The first page of the survey, ‘Welcome!’, consisted of a welcome note with a brief definition of an Orphan Work and a notice about how the data from the survey would be used and who could be contacted in case respondents had any questions about data protection. The second page, ‘Demographic Data’, was not compulsory but collected information about job titles of respondents and their co-workers. As we planned to offer a follow-up option after the end of the survey, we did offer a box to include an email address, but it was not mandatory to provide this. The third page, ‘What is the UK Orphan Works Licensing Scheme?’, is where piping began, with those who had not heard of the Scheme being removed to a ‘Thank You’ page, and those who had heard of the Scheme continuing on with the survey. This was to weed out potential trolls, bots, and anyone who had taken the Survey to be helpful but whose answers might have skewed our results. The fourth page for respondents who had heard of the Scheme, entitled ‘Your Experience with Digital Work’, asked questions about the scope of digital projects that a respondent might be familiar with, and whether any of those projects included the use of Orphan Works. The sixth page, ‘Your Institution and Orphan Works’, asked respondents to describe their knowledge about their institution’s use of Orphan Works, as well as any potential policy documents developed by their institution regarding Orphan Works. As some respondents were using Orphan Works individually, this section was not compulsory. The next page, ‘Have you used the Orphan Works Licensing Scheme to Apply for a License?’, also included piping. If the answer to the question was ‘Yes,’ the respondent was directed to page 8, ‘The Intellectual Property Office Orphan Works Implementation’. If the respondent answered ‘No’, they were directed to page 9, ‘I Know about the Orphan Works Licensing Scheme, But My Institution/I have not used it’. After this, both sets of respondents were directed to page 10, ‘May we contact you for further input?’, which asked whether we could follow-up with respondents and whether they had anything else to tell us about the Orphan Works Licensing Scheme that we had not covered in our questions. The final page was a ‘thank you!’ page and again gave our contact information in case respondents had further questions about the survey itself or about how their data would be used.

Analysis and Results of our Survey

In total we received 59 responses to our survey. As the pool of potential respondents who had actually used the OWLS was so small (35 potential institutions/companies/users), the volume of feedback we received gives a good picture of a niche user community, as well as providing us with some insight from those who had not used the Scheme for specific reasons.

After collecting the responses, we exported the data from SurveyMonkey to a .csv file, deleting personal information including names, IP addresses and email addresses prior to analysis. We assigned the respondents random numbered IDs so that the data, when shared or requested, would not contain personal information beyond the job title, institutional affiliation, and geographic location (when provided). After respondents were randomized, the individual responses were coded and separated based on two initially identified types of responses: those who had heard of the Scheme and those who had not. We formatted the data with as little intervention as possible, though some spelling and geographic markers were standardized.

Because we recognize that job titles do not always portray the full gamut of an individual’s skills, we asked respondents to describe how they would characterize themselves. Respondents were encouraged to choose as many roles as they felt applied to them. We wanted to understand what types of jobs the Orphan Works problem was affecting and to provide evidence that those working on or familiar with the problem were a more wide-ranging group than just copyright specialists.

Choosing multiple roles that suited their responsibilities, 32% of our respondents categorized themselves as Scholars, followed by 27% opting for Professor/Lecturer/Reader/Tutor. The titles Librarian and Editor (Textual/Scholarly) were split evenly in responses at 20%. However, the majority of the 59 respondents, 34%, chose the ‘Other’ category, self-identifying as ‘trainer and consultant in digital libraries and digital preservation’, ‘museum worker’, ‘copyright/licensing professional’, ‘lawyer’, ‘administrator’, ‘consultant’, ‘writer’, and ‘descendant of an artist’. As our demographic data section was not compulsory, we did not collect more specific information about what portion of the respondents represented individual GLAM institutions. In some cases, these respondents were working on behalf of an organization when they applied to the Scheme and in others they were working for a small business or for their own individual use of an Orphan Work. The lack of designers or artists from this category is telling, regarding the minimal take up of this Scheme in the creative industries (although this may be due to the reach of our survey).

In a question that allowed for multiple responses, we asked respondents to characterize their colleagues, so we could understand how people interested in Orphan Works are collaborating with other members of the cultural heritage sector, or indeed other sectors. The same job titles were included from the previous list, respondents could choose more than one characterization, and responses indicated that over half (54%) of our respondents characterized their colleagues as Professors/Lecturers/Readers/Tutors, followed closely by Librarians (50%). Of the ‘Other’ responses, which made up 20% of the total number, respondents cited their colleagues as ‘academic professional support’, ‘museum workers’, ‘loose group of historic computing enthusiasts’, ‘attorneys’, ‘press and policy officers’, and ‘normal people!’.

Over 60% of our respondents stated that they or their institution had not applied for a license using the Scheme. Of those respondents who indicated that they or their institution had applied for a license, 14% responded that they had applied for a commercial license while 26% replied that they had applied for a non-commercial license. ‘I don’t know’ garnered 43% of the responses, and the ‘Other’ option was used by 17% of respondents. Respondents 44 and 24, respectively, replied that they were ‘Still in the process of discussions’ and ‘would suspect it would be non-commercial if we did apply’.

When asked if they had performed a diligent search prior to applying for a license with the Scheme, 57% of respondents replied ‘Yes’ and 43% used the ‘Other’ box to write in ‘I don’t know’. Respondent 24: ‘I certainly have done already re photographs trying to get together as much information as I can re possible copyright holders, have largely been successful but there are still some gaps. Correspondence [h]as proved to be even more of a headache’.

We then asked respondents to describe the steps they or their institution had taken for a diligent search prior to applying for a license with the Scheme. Responses included: Respondent 58:

‘consulted the author’s will as well as the will of the person to whom copyright was bequeathed’; Respondent 51: ‘I don’t know’; Respondent 46: ‘We followed the IPO published guidance’; and Respondent 24:

[p]ersonally, as far as photos were concerned, I checked with the National Portrait Gallery and the Imperial War Museum (many of our photos overlap with their holdings) to see if they had any further information. Both were very helpful in redirecting me to rights holders. I googled names, checked old telephone directories, even used 192.com [a paid service to look up individuals] to try and track down descendants of rights holders. Also contacted newspapers/journals where it looked as though photos had originally been produced for them. Occasionally I have used a photo online asking for help in tracing the rights holder, this has generally been unsuccessful.

The respondents were asked if they or their institution had any technical difficulties with the Scheme’s site, Respondent 35 commented that ‘it looked very complicated, I didn’t have the will to go through the process’.

Open-text answers about reasons for not using the scheme were instructive. Responses included: Respondent 49: ‘Apart from the cost of registration, the due diligence requirements would be too expensive for us to undertake apart from the exceptional cases’; Respondent 41: ‘It will only be applicable to works created in the UK that we hold in our collection’; Respondent 22: ‘Took a view that the risk was not in line with the admin and time involved for our archival material’. Respondent 35:

The usefulness of a given potential OW item has so far not justified the amount of work involved contacting all the various agencies etc to get registration. Taking a risk based approach we have published a few things in the meantime however.

We asked how the scheme could be improved in order to make it more enticing to potential applicants. Responses included: Respondent 51: ‘Please, fees to be reviewed’; Respondent 48:

No direct experience of applying for a license yet- however diligent search may be too onerous- or applicants may be being too diligent (cautious). Generally fees are too high. A lot of Orphan works are orphan because of their limited commercial viability so rights holders had no interest in investing in chain of title; renewal of underlying rights etc- fees should reflect this. Fees also wouldn’t reflect other investments/costs of publishing a work that are often part of any licensing deal with right holder (along with recoupment of costs or share of costs;

Respondent 38: ‘Needs to be faster, larger users should have an account and their due diligent work should be approved by the IPO so if the organization applies the due diligence is already done’; Respondent 35: ‘Currently the amount of effort involved in checking all the agencies etc is a major obstacle – cutting the number would make it easier’; Respondent 27:

Seems to me that it’s cheaper and more cost effective for institutions who have done the diligent search not to register. Extra fees for very little benefit. More effective to take fees and put them to one side to pay copyright holder directly when they come forward. So changing the fee structure might help. Also license length is too short- only 7 years!;

Respondent 24: ‘Seven year [license] and then hav[ing] to reapply seems a bit silly. Not enough publicity about the Scheme, or where to find further information’.

When asked if there was anything they wanted to tell us about the Scheme that hadn’t been covered already, respondents wrote: Respondent 49: ‘We are more likely to use the EU process since it doesn’t require payment and relies on self-certification for due diligence’; Respondent 38: ‘I think big users such as the BBC who could potentially use [the Scheme] a lot will be put off by the one at a time approach’; Respondent 27: ‘It’s not the license fee that’s the issue – it’s the application fee. It’s only worthwhile if the product is commercial in nature. EU and UK might be better served with a copyright exception as supported by the ICOM [International Council of Museums]’; Respondent 22: ‘Not sure who it is really aimed at or why – do small archives really need to worry about this? Or even large ones? Or should this really just be aimed at high profile, potential high value Orphan Works? Seems to be just causing confusion and worry’.

These responses provide us with a baseline for understanding the concerns of users and potential users of the Orphan Works Licensing Scheme. Respondents expressed that the Orphan Works Licensing Scheme had not dispelled concerns about the cost and time of a diligent search, stating that it instead created concerns about the duration and cost of non-commercial licenses geared toward aiding the online dissemination of academically valuable materials in the cultural heritage community. In fact, our research shows that the Scheme, rather than having improved access to Orphan Works, has actually stymied efforts by cultural heritage organizations to engage in digitization of Orphan Works on a massive scale. Perhaps the most telling evidence for this is that to date, large cultural heritage organizations such as the BBC, the National Libraries of Wales and Scotland, respectively, the British Library, and the Imperial War Museum have not applied for licenses from the UK Orphan Works Licensing Scheme. This is notable given that these institutions have large collections of Orphan Works. Additionally, given the vast volumes of Orphan Works material in the UK, and the potential user communities willing to access these resources, the fact that only 35 separate researchers, companies, and institutions had used the Scheme to register and apply for licenses of Orphan Works in the first fourteen months of its operation indicates that is it not the transformational approach needed to open up cultural heritage assets across the sector.

Brexit, the EU copyright exception, and the UK Orphan Works Licensing Scheme

In the intervening year since our survey was closed, there has been little change in the legislation of Orphan Works, but there has been significant socio-political upheaval. If the confusion surrounding Orphan Works were not headache-inducing enough, the prospect of Brexit also brings uncertainty for cultural heritage institutions in the UK. In October 2018, the UK Libraries and Archives Copyright Alliance (LACA) published a post on their website arguing that:

as a likely consequence of the UK leaving the EU, the UK is at risk of losing a hugely beneficial ‘exception to copyright’ that allows cultural organi(s)ations to digiti(s)e and make available online ‘orphan works’ from their collections. The Government has said that in the case of a ‘no deal’ Brexit this helpful exception will be removed from UK Law (2018: n.p.).

The citation that LACA refers to in the above quote comes directly from the UK Government’s Fact Sheet on a ‘no deal’ Brexit: ‘UK-based Cultural Heritage Institutions that make works available online in the EEA under the exception may be infringing copyright’ (UK Department for Business, Energy & Industrial Strategy, 2018: n.p.). As the digital landscape means that making a work available online in one country means that it is available anywhere, this is essentially a death sentence for digital Orphan Works collections as we understand them in this legal framework.

In the flyer attached to the post, LACA urges relevant stakeholders to perform a few steps in order to save the provision of the copyright exception in UK law and, by extension, their collections. However, using the EU copyright exception, which LACA deems an ‘easy’ process, is anything but: the advice that LACA provides still does not account for the difficulty or expense of performing a diligent search, particularly on unpublished items such as manuscripts, letters, films that have not been broadcast, and/or private photograph collections.

The flyer (Figure 7) points stakeholders to links for three potential starting points that provide diligent search guidance: the EnDOW project,5CopyrightUser.org,6 and the UK Intellectual Property Office’s own diligent search guidelines.7 The link to the EnDOW project provides specific guidance for published works, with further links to country-specific databases, the CopyrightUser.org link provides a basic overview of what an Orphan Work is and how the OWLS works, and the link to the UK IPO gives stakeholders a set of diligent search guidelines to focus on for different materials, all of which are geared toward licensing with the OWLS. In essence, only two of the three links provide any relevant guidelines for diligent search, and they are both focused on diligent search for published works.

Figure 7 

The Libraries and Archives Copyright Alliance ‘Use It or Lose It’ campaign flyer (LACA, 2018).

Aside from the mischaracterization of the time and effort needed to perform a diligent search, the advice given by LACA first asks interested parties to go to the self-assessment tool on Gov.uk to ‘check that you and your “orphan work” collections meet the criteria’ to use the EU copyright exception’ (LACA, 2018: n.p.). Again, this presents a practical and a more nuanced problem. The UK IPO released the self-assessment tool simultaneously with the UK Orphan Works Licensing Scheme for users to determine whether they qualify for the EU copyright exception and therefore do not have to use the OWLS to license their works and upload them in the UK Orphan Works Register.

The problem stems from the fact that the self-assessment tool assumes that a potential user has already conducted a diligent search for rights holders — meaning that for LACA to suggest diligent search as a second step on their flyer, after taking the self-assessment test, is incorrect. However, this error by LACA is hardly surprising, given that a potential user only realizes a diligent search is needed to properly take the self-assessment tool during the last two questions of the assessment, and even these are implicit suggestions, and expose the user to potential liability.

The first six questions of the self-assessment (see Figures 8 and 9 above) are innocuous and require answers that would not indicate that a diligent search has already been performed. Of the eight questions in the self-assessment quiz, question seven asks ‘As far as you are aware, has the work been published, broadcast or otherwise distributed within the EU?’ (UK IPO, 2018: n.p.). Unless significant metadata about an individual item or collection were already available, which in the case of Orphan Works is hard to gauge on an item by item (or collection by collection) basis, a user would need to perform a diligent search in order to answer this question.

Figure 8 

The UK IPO Self-Assessment Tool, Question 1. Choosing any of the options besides ‘None of the above’ allows a user to move on to the next question. Choosing ‘None of the above’ generates a response of ‘Your organisation does not appear to be eligible to use the work/performance under the exception implementing the EU Directive on certain permitted use of orphan works, however you may be eligible to apply using the UK Orphan Works Licensing Scheme,’ with a link to the Register (UK IPO, 2018b).

Figure 9 

The UK IPO Self-Assessment Tool, Questions one through six. All questions offer either a ‘Yes’ or ‘No’ answer choice. (UK IPO, 2018: n.p.). Choosing ‘No’ for questions one through five generates the same message as in Figure 5 above: ‘Your organisation does not appear to be eligible to use the work/performance under the exception implementing the EU Directive on certain permitted use of orphan works, however you may be eligible to apply using the UK Orphan Works Licensing Scheme’ (UK IPO, 2018: n.p.).

The three answer options are ‘Yes,’ ‘No,’ and ‘Don’t Know’. Responding with any of the three answers makes no overall change to the outcome of the assessment; if question one is answered with one of the organizations listed, and if questions two through six are answered with a ‘Yes’, then the answer to question seven with any of the answers makes no difference to the overall assessment. In this scenario, it is the final question which really determines whether a potential user is able to qualify for the EU copyright exception. The final question (eight, seen above in Figure 10) of the self-assessment tool asks users to declare whether ‘As far as you are aware, would the rights holder allow you to reproduce the work for the purposes specified in the Directive?’ (UK IPO, 2018: n.p.). This question provides only a ‘Yes’ or ‘No’ option. No information is given as to the purposes specified in the Directive, so it is assumed that a user will already be familiar with these purposes. By providing an answer to this question, not only is there an assumption that a diligent search has, to some extent, been performed, but the UK IPO is implicitly asking an individual staff member – on behalf of an organization – to make a value judgment from the point of view of a rights holder (or rights holders) whom they do not know and cannot locate. A ‘No’ answer brings up the following message: ‘It appears that you would be unable to use this work under the Directive’ (UK IPO, 2018: n.p.). A ‘Yes’ answer to question eight, regardless of the answer to question seven, generates a message stating:

Your organisation appears to be eligible to use the exception under the EU Directive. You will need to register further details relating to the diligent search on the [EU] [o]rphan [w]orks [d]atabase on the Office for Harmonisation in the Internal Market (OHIM) website (UK IPO, 2018: n.p.).

Figure 10 

UK IPO Self-Assessment Tool, Questions one through eight. Note that question seven shows an answer of ‘Don’t Know’ and the authors were still able to continue to question eight (UK IPO, 2018: n.p.).

Answering ‘yes,’ when a rights holder is unknown or cannot be located seems to expose cultural heritage organizations to potential risk. However, the self-assessment tool does not require a login or any verification of the type of organization with which the user is associated. The assessment result is not a legally binding agreement, but does indicate that by the UK IPO’s standards, and the user’s understanding of those standards as they are laid out in the assessment, the result will be sound and legally justified. Alternatively, it can be supposed that the UK IPO is leaving the decision, and therefore the liability, entirely in the hands of the user: if the user asserts that the hypothetical rights holder would be happy for the work to be used, then the user can register the work in the EU Orphan Works Database and digitize it, hoping that their decision would be in line with the hypothetical rights holder’s wishes.

What the LACA flyer doesn’t mention is that if a ‘no deal’ Brexit does occur, there is no amount of ‘collection saving’ that will change the basic facts of Article 50 of the Lisbon Treaty, which was drafted in 2007 and went into force in 2009 (European Parliament and the Council of the European Union, 2008). The Article details the knock-on effects of a Member Country leaving the EU. Section 3 of Article 50 particularly states that, barring an extension agreed upon by all EU Member States:

the Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period’ (The Lisbon Treaty, Article 50 Section 3, 2009).

In other words, all the benefits and protections that the UK receives from EU treaties, including the copyright exception for non-commercial use of Orphan Works which allows cultural heritage organizations to register their works in the EU Orphan Works Database, will cease to exist (BBC, 2017: n.p.). There is not much that can be done if the EU Member States do not agree to an extension. Certainly LACA’s advice to utilize the EU Orphan Works Database as much as possible is good in theory. If the UK and the EU do not meet a reciprocal agreement that includes protections for directives that are already in force in UK law, the time and labor intensive process may have been in vain, and the UK GLAM sector’s ability to use the copyright exception and the EU Orphan Works Database will be called into question. This leaves cultural heritage organizations in a difficult position, not just with the future of their Orphan Works collections, but also with any previous collections that have been made available online after using the EU Orphan Works Database.

Conclusion

At time of writing, four years have now passed since the UK Orphan Works Licensing Scheme began accepting applications for commercial and non-commercial use of Orphan Works. As of October 2018, 144 licenses have been granted of a total of 877 items. That is an average of 18 works per month over the 48 months that the Scheme has been operational, a far cry from the UK IPO press release dated 29 October 2014, entitled ‘UK opens access to 91 million Orphan Works’ (2014). This perfectly illustrates what the impact assessment commissioned by the UK IPO predicted: that such a bureaucratic system requiring individual licensing and data entry would be ‘very little used’ (2012: 6).

Aside from conducting our own survey on Orphan Works, we also analyzed a 2015 survey carried out by the UK Intellectual Property Office itself. With 19 respondents from a pool of 80 recipients, the survey is also qualitative (UK IPO, 2015a). The UK IPO survey garnered similar responses as our own, leading to similar conclusions as ours. The survey alerted us to one very important aspect of the UK IPO’s Scheme and how the Government plans to improve it: because of potential difficulties caused by impact assessments and business reports if a digital system has gone ‘live’, the UK IPO has intentionally kept the Scheme in beta so as to implement changes without having to undergo these types of reviews. This answers a question we had when writing this article: why, after almost four years, would an online system still be in beta? It is because potential changes would be made faster by doing so, with little to no oversight (UK IPO, 2015a). As the UK Orphan Works Licensing Scheme has been intentionally kept in beta mode so as to implement improvements without necessitating a business impact assessment, we believe our commentary adds significant value. As stakeholders, users of the scheme (Terras, 2015), and researchers, our recommendations for improvement could potentially impact the cultural heritage sector across the UK. Our recommendations are in line with those provided in the UK IPO’s previous survey (2015), as well as that carried out by Naomi Korn (2009) and the responses to the Gowers Review (2006).

Currently the UK Orphan Works Licensing Scheme does not adequately address the concerns of the cultural heritage sector. The main issues with the Scheme are that the costs, including license and application fees, VAT, and potential Copyright Tribunal appeals, all indicate significant costs for potential users. Technically the cost seems reasonable: 10 pence per work for a non-commercial license and 30 works per license at 80 pounds sterling per license, plus a VAT charge per license which is kept by the UK IPO to help fund the cost of running the Scheme and for which potential rights holders can claim if they come forward and prove they have copyright for a work on the Register. However, when these hypothetical costs are extrapolated to cover the vast number of works held in memory institutions – sometimes millions of works – the Scheme does not work at scale. Beyond the upfront fees, there are also the added costs of labor spent undertaking due diligence. The original problems identified by multiple impact and scoping reports of the past fifteen years still remain. It should also be noted that the cost structures put in place for commercially licensing material via this Scheme are prohibitively high and ambiguously structured. Indeed, a thorough explanation of the cost structure for commercial vs non-commercial licensing is not provided or explained on the UK Orphan Works Register page of the GOV.UK website. The Scheme can therefore put further financial barriers in place for individuals or smaller companies wishing to create products using licensed Orphan Works material.

Our primary recommendation (novel from those previously published) is that instead of pursuing the time-consuming, costly, and short-lived licenses of the UK Orphan Works Licensing Scheme, the cultural heritage sector in the UK should eschew it altogether and take a risk management approach instead. Our reasoning for this suggestion is that the Scheme itself does not cater to larger-scale digitization and related licensing, nor does it feasibly offer the opportunity to allow creative reuse of content or cost-effective monetization of collections. Further, the UK Scheme does not cover licenses beyond the UK, thus making them potentially meaningless in the digital realm, where information moves seamlessly across geographical borders. Memory institutions could potentially use the EU Directive Orphan Works Database rather than the UK Scheme to register non-commercial use of Orphan Works, and this is more desirable given that diligent search is recorded using an honor system and fees for rights holders coming forward would hypothetically be agreed between the rights holder and the cultural heritage institution. However, the EU Directive could potentially be unavailable to the UK Glam sector in the near future with Brexit on the horizon.

Attempting to ascribe the same set of guidelines for the average individual user as to the cultural heritage organization in the UK does not work to scale, nor does it allow for the fact that the sector would have much more need of such a scheme than one user or small company. Therefore, the best-case scenario for UK cultural heritage organizations, given the problems with the UK Orphan Works Licensing Scheme and the uncertain future of the EU Orphan Works Database, is to use their own initiative to digitize and publish Orphan Works with the appropriate take down notices (if the copyright holder is subsequently identified and requires removal of the work from an online collection) and a rigorous and well-documented blanket policy to accompany each known Orphan Work. Of course, the sacred trust of custodianship is important, and no GLAM wants to expose itself to potential risk. At the institutional level, perhaps some GLAMs in the UK are designing Orphan Works strategies behind the scenes, and if so, this is a good and necessary thing. When such information is provided so as to potentially attract rights holders, cultural heritage organizations can rest assured that they have acted responsibly and also have allowed the rights holders to be responsible for producing burden of proof that they own the rights to an item, rather than the other way around. In the case of the creative industries and companies that are attempting to monetize products based on Orphan Works, this may seem as though there is advantage taken of potential rights holders. However, for the cultural heritage sector, which is largely supported by public funds, and for which the purpose of such organizations is the dissemination of knowledge and culture to a wide range of users, this makes sense. Rather than being penalized with heavy costs, time consuming processes, and long diligent search waits (for responses from organizations and individuals), GLAMs can instead focus on collection building and fulfilling their public mission of access. The burden of the risk in the commercial use of an Orphan Work item would then fall to the individual user or company – although clear guidance and explanation from the host institution would be needed to explain the situation.

A thorough and wide-ranging study surveying cultural heritage institutions and members of the private sector across the UK in light of the implementation of the EU Directive 2012/28/EU would provide ample empirical evidence for the reasons why the cultural heritage sector is eschewing the UK Scheme. As yet, the only in-depth study detailing the implementation of the UK Scheme focuses on the process and the databases needed to conduct a diligent search (Favale, Schroff & Bertoni, 2016). It would provide more support for the claims made by users and potential users in our own study, and the study performed in 2015 by the UK Intellectual Property Office. The EU IPO is also currently conducting a survey and collecting responses for its Database, and the results of this survey may shed more light on how one harmonized database is preferable to individual implementations across Member States (EU IPO, 2018).

Easing the burden of diligent search and license fees for institutions would cut down on resources expended (the original work done by GLAMs to perform lengthy diligent search is then repeated by the UK IPO, which checks said due diligence, effectively performing it twice). It would also realign the mission of cultural heritage institutions with what is possible in the digital realm: namely the mass digitization of collections and their distribution across the globe, in a way which allows and facilitates different types of reuse in the creative industries as well as for research. Re-contextualizing physical collections through the use of digital technology helps to bridge connections and vastly change the way we see items. Especially in the case of publicly funded GLAMS, depriving the public of access to the collections they are paying to maintain with their taxes is nonsensical. The commercial licensing of works owned by others is still necessary in the copyright culture in which we live: however, the overall point is that cultural heritage organizations should not be penalized, nor should the public or potential users, because copyright law cannot keep pace with the opportunities offered by the digital realm.

Most importantly, what must occur is a fundamental reorganization of thinking on the part of the UK Government about GLAMs’ relationship to Orphan Works, particularly as regards the amount of time and money spent performing diligent searches for rights holders. Risk appetite at memory institutions with large collections is historically low, but education about the risks involved through more research on previous cases of rights holders coming forward to claim ownership of cultural heritage items or collections could sway the tide. The potential to be gained through research or creative projects using these items could favor librarians, archivists, and other scholars charged with preserving and disseminating Orphan Works. It would also privilege users taking, making, and not-lawbreaking in order to utilize the vast resources at hand both for traditional research approaches, but also in more creative ways. False barriers are currently in place that keep users from engaging digitally with Orphan Works materials, whether for routine or more creative access.

However, change will not be brought by problematically implemented policies such as the UK Orphan Works Licensing Scheme. Rather, it will come through librarians, archivists, and other stakeholders knowing that they have adequate support from their institutional boards and CEOs so that they might tackle the Orphan Works issue themselves via fair dealing legislation and judicious use of take down policies, in order to increase access for users and build new relationships with digital communities in the online age, and with brave users willing to navigate and question currently confusing frameworks. Regardless of any upcoming changes at the national governmental level, a risk management approach at the institutional level toward Orphan Works and their dissemination to the research and creative user community would serve both the public good and test the idea that the digital landscape is a space for the democratization of culture, leading to innovative and potentially discipline-shaping research. The further opening of cultural heritage content to both the research and creative industries, and individual users wishing to play with Orphan Works cultural heritage content, could also be transformational in the way that both institutions and users think about digitization and potential use cases for digital cultural heritage content: but this can only be done if the current copyright frameworks which put barriers in the way of use and innovation are reconceived. In the GLAM sector, perhaps no one wants to (publicly) be the one who moves away from a dissatisfying but familiar paradigm. Nevertheless, with the current frustrating copyright situation behind and the radical possibilities of research and remixing of Orphan Works ahead, the first step must be taken.