This article is from the June 2003 issue of Update. 

Digital media have a relatively short life expectancy. Preserving digital information for the long term presents many problems,1 the biggest of which arguably come from changes in coding, formats, software, operating systems and hardware that can render digital material unreadable.

The complex nature of much digital information can also present difficulties. There is a plethora of formatting standards for different types of digital information, which may not be implemented uniformly or may change rapidly. Complex digital material may also be dependent on software for search and retrieval and other functionality. The ephemeral nature of much online information also has implications for digital preservation. The rights in the content and any associated software may belong to a number of different individuals or organisations. Digital information may be surrounded by technology designed to protect it from unauthorised copying and redistribution, which may also hinder preservation.

For traditional media, libraries acquire and physically own a discrete information object. In the digital world, the model is paying for access to information held remotely. If libraries do not own digital material, they cannot preserve it. Publishers may or may not have a commitment to preserving their own information, depending to an extent on what type of publisher it is and what its mission is.

Table 1. Summary of copying implications of digital preservation

Redundancy, migration and emulation
Redundancy is part of traditional preservation management and creating backups is part of good security practice for digital data. Other strategies include migration and emulation and variations on these.

Migration is 'the periodic transfer of digital material from one hardware/software configuration to another, or from one generation of computer technology to a subsequent generation.'2 While refreshment and media migration ensure that a reliable bit stream of the digital object is maintained, migration strategies may result in a loss of data, or changes in functionality and the look and feel of the digital object. Migration on demand may minimise this because 'all the migration from an obsolete format to an appropriate current day format happens at the point of request'.3

Emulation involves a current or future technological platform mimicking an older or obsolete platform. It can be at the hardware or software level. Emulation will also involve refreshing or migrating media over time. There is also likely to be some sort of encapsulation of content, software and metadata. Metadata could include a specification for an emulator.4 Alternatively, emulators could be designed and then updated over time.5 Re-engineering of software may also be necessary to specify or create emulators.

Lorie6 suggests that, while emulation could mimic obsolete technologies to allow display of the original, it will not allow future users to manipulate the information. The aim of the Universal Virtual Computer (UVC) idea is to develop '…virtual machines that can execute essential functions on a variety of platforms'.7 Information is converted to a 'preservation format' that will allow a program to extract data from the bit stream and present it in an understandable form. The program runs on a UVC, a program comprising low-level instructions that can, in theory, be interpreted to run on any computer in the future.8

All these potential strategies have something in common. In order to preserve digital information, it has to be copied and perhaps changed in some way. For libraries to be able to do this they need to own the information.

AHRB project
It is not clear whether copyright legislation and licensed access to digital content threaten the ability of libraries to provide long-term access to that content. The aim of the Copyright and Licensing for Digital Preservation Project, sponsored by the Arts and Humanities Research Board, is to investigate these issues and to suggest ways in which any problems can be overcome. The research is focusing on the UK, but the findings are likely to be of wider interest, and the project is collecting information on the situation in other countries.

The specific aims of the project are to:

• assess whether the provisions of UK copyright legislation meet the digital preservation needs of UK national, academic, public and special libraries, and how copyright legislation in other countries addresses this issue;
• investigate to what extent licensed access to digital material in libraries takes account of preservation needs and how publishers and information providers plan to achieve the provision of ongoing access to digital material for libraries;
• make recommendations for amendments to UK legislation and model licences for long-term access if appropriate;
• make recommendations on how legislators, information providers and libraries can work together to ensure long-term access to digital information.

The research involves a review of the literature, questionnaire surveys, focus groups and interviews. This article reports the initial findings of the literature review.

Copyright exceptions
Current legislation in the UK provides limited exceptions to copyright. One of these exceptions is the 'library privileges', including copying for purposes of preservation or replacement.9 This exception is very narrowly defined; it permits a librarian or archivist of a prescribed library (school, university, FE, public and government) or archive to make a copy from any item in the permanent collection in order to preserve or replace that item, providing that the prescribed conditions are complied with.

Almost all digital information is made available via one form of database or another. An EU directive on databases was issued in 1996 and a statutory instrument was subsequently passed in the UK to protect databases.10 Databases enjoy double protection — the database or sui generis right, and copyright. The database right applies where there has been a substantial investment in obtaining, verifying or presenting the contents. The right prevents the unauthorised extraction and re-use of material from a database, whether it enjoys copyright or not. As a result, both the copyright residing in the structure of the database and the database right restrict the transference of databases to another medium. However, it is not an infringement if a person who has a right (by licence or otherwise) to use the database (databases as literary works) exercises their right to access the database and to use its contents by whatever means necessary. Any term in the contract or licence that prohibits this is irrelevant. It is not clear what this means for digital preservation.

An EU copyright directive11 designed to harmonise various aspects of copyright law among member states will be implemented soon in the UK. It recommends that exceptions and limitations should be defined more harmoniously. However, it does not make this compulsory and we do not know yet how the UK will implement the directive. Some of the provisions are of potential interest. For example, Article 6 instructs member states to provide adequate legal protection against the circumvention of technological measures that provide legal protection for copyrighted works. However, Section 4 of Article 6 states:

'… in the absence of voluntary measures taken by rightholders, including agreements between rightholders and other parties concerned, Member States shall take appropriate measures to ensure that rightholders make available to the beneficiary of an exception or limitation ... the means of benefiting from that exception or limitation ... where that beneficiary has legal access to the protected work or subject-matter concerned.'

Depending on interpretation, this may allow preservation agencies the right to circumvent protection measures in order to preserve the material, once they have met the requirements of the exception.

The only type of copying that seems to be clearly allowed under the UK copyright law preservation exception is the first act of media refreshment or migration. If digital information is copied to paper or microform, then users would be able to access this version because use requires no further copying.

The creation of multiple copies for the purposes of redundancy is also not allowed under the law. Database provisions need to be clarified. Copying of material that libraries can access, but do not own, does not seem to be allowed under library privileges. When the EC directive on copyright is enacted, libraries may be able to circumnavigate copyright protection to exercise their privileges. Since the preservation exception does not allow much copying for preservation purposes, this is not particularly useful.

Extending legal deposit
If libraries are to ensure the preservation of digital information, they can either rely on publishers to take on this responsibility or physically acquire the information. Extending legal deposit12 to cover digital publications may get over the problem of access to, rather than ownership of, digital information, but only to a limited extent because legal deposit only covers the national digital output. Even if publishers deposit their digital information, the question of whether deposit libraries would be able to carry out preservation-related copying or whether further provision will be have to be made through copyright law would need to be clarified.

The proposed new UK legal deposit legislation acknowledges this by providing for amendments to rights legislation so that 'copyright, publication right, database right and any similar right' is not infringed by deposit libraries who need to copy for preservation or the provision of access or uses making temporary copies for access purposes.13 However, this still leaves the question of whether other types of libraries can legally preserve digital material they physically own.

Pulling information from publisher sites will involve copying. Depending on how it is implemented, the new UK legal deposit legislation, and any accompanying amendments to rights legislation, may deal with harvesting web material 'published' in the UK.

What are other countries doing?
Other countries are also addressing this issue. The Norwegian National Library has embarked on a project that will, among other things, explore the legal issues of collecting online material. The Swedish government issued a decree authorising the Royal Library not only to collect Swedish websites on the internet but also to allow the public access within the library premises.14 The French government has adopted a law requiring every French website to be archived,15 allowing preservation organisations to harvest material at regular intervals. The Joint Information Systems Committee and the Wellcome Trust recently commissioned a study on legal issues related to internet archiving in the UK, EU the US and Australia.16

There are some examples of countries that have legislation that looks more amenable to copying for digital preservation. In the US, the Digital Millennium Copyright Act (DMCA) expressly allows authorised institutions to make up to three digital preservation copies of an eligible copyrighted work. It allows the institution to loan those copies to other institutions and permits preservation, including by digital means, when the existing format in which the work has been stored becomes obsolete.17 The Canadian Copyright Act allows electronic publications to be converted to a contemporary format if necessary for preservation purposes. The work must be in the library's permanent collection and the copy must be made in order to maintain the collection.18

Licences and perpetual access
If libraries do not have the legal right to copy for preservation purposes, one approach they could take is to ask rights-holders for permission. However, clearing rights is likely to be resource-intensive and difficult.

Model licences could reduce the burden on libraries and publishers. The Jisc Model Licence for Journals19 allows the licensee to make backup copies of the licensed material in order to make them accessible locally. The publisher undertakes to provide access to the archive of subscribed material once a subscription ends. However, this relies on other players taking responsibility for and ensuring preservation. Even if guarantees are given, organisations go out of business or just may not have the resources to do the job. Alternatively, the publisher may provide the former licensee with an archival copy of the material that was subscribed to, in a mutually agreed format. However, the delivery of the archival copy will not ensure long-term access to the information unless the licensee has permission to copy for preservation purposes. Jisc has recently funded a one-year study to explore, with publishers and other stakeholders, archiving and access provisions and to evaluate past licences and future options for archiving of licensed e-journals and access arrangements.20An alternative to individual libraries seeking permissions is some form of central rights clearance operation. We are not aware of any such schemes in place for preservation copying of digital material in the UK or elsewhere.

Another possible alternative to clearing rights could be including rights information in metadata associated with digital material.

The OCLC/RLG Working Group on Preservation Metadata21 has developed a preservation metadata framework that includes a rights-related element. This needs further development and it is not clear how it would be implemented and populated. The metadata would have to come from rights-holders, so it would have to be incorporated into publisher metadata schemes, such as the Online Information exchange (Onix).22 However, publishers may not have any incentive to populate preservation rights elements.

Conclusions
It is not clear whether UK law provides for copying for digital preservation. Copyright law could be amended to allow the copying required for digital preservation; other countries could provide examples to be followed in the UK. Access rather than ownership models also present problems for libraries in that they have to rely on others to preserve digital information. While physical deposit of digital information with libraries is the only option that gives libraries control over preservation, without the legal right to copy, this control is only partial. Libraries and publishers will have to work together to agree how the preservation of digital information can be managed.

The next step of this research will investigate the needs and views of these stakeholders. The data gathered will be used to come up with more potential solutions and assess what actions need to be taken by libraries, publishers, legislators and others.

References

1 M. Deegan and S. Tanner. 'The digital dark ages.' Library & Information Update, 1 (2) May 2002, pp. 42-3.
2 D. Waters and J. Garrett. Preserving Digital Information: report of the Task Force on Archiving of Digital Information commissioned by the Commission on Preservation and Access and the Research Libraries Group. Washington, DC: Commission on Preservation and Access, 1996.
3 Cedars Project. Cedars Guide to Digital Preservation Strategies (www.leeds.ac.uk/cedars/guideto/dpstrategies/
dpstrategies.html).
4 J. Rothenberg. Avoiding Technological Quicksand: finding a viable technical foundation for digital preservation. A report to the Council on Library and Information Resources. CLIR, 1999 (www.clir.org/pubs/reports/rothenberg/contents.html).
5 D. Holdsworth and P. Wheatley. 'Emulation, preservation, and abstraction.' RLG DigiNews [online], 5(4) (www.rlg.org/preserv/diginews/diginews5-4.html).
6 R. Lorie. 'A project on preservation of digital data.' RLG DigiNews, 5(3) 2001 (www.rlg.org/preserv/diginews/diginews5-3.html#feature2).
7 K. Thibodeau. 'Building the archives of the future: advances in preserving electronic records at the national archives and records administration.' D-Lib Magazine, 7(2) 2001 (www.dlib.org/dlib/february01/thibodeau/02thibodeau
.html).
8 Ibid.
9 Copyright, Designs and Patents Act 1988, Ch. 48, 36 & 37 Elizabeth II, ss. 37-42. HMSO, 1988.
10 Copyright Rights in Databases: the Copyright and Rights in Databases Regulations 1997, 45 & 46 Elizabeth II. HMSO, 1997.
11 European Commission. Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society, 2001.
12 'Bill to save e-heritage.' Library & Information Update, 2(3) March 2003, p. 5.
13 Legal Deposit Libraries Bill. Bill 26 53/2, 7.
14 'New decree for Kulturarw3, 2002:287'. 2002 (www.kb.se/Info/Pressmed/Arkiv/2002/020605_eng.htm)
15 'Entire French web to be archived', 2001 (www.europemedia.net/shownews.asp?ArticleID=4075).
16 A. Charlesworth. Legal issues relating to the archiving of internet resources in the UK, EU, USA and Australia: a study undertaken for the Jisc and Wellcome Trust. Jisc; Wellcome Trust, 2003.
17 Digital Millennium Copyright Act, Title IV (DMCA) 1998 (www.loc.gov/copyright/legislation/dmca.pdf).
18 Copyright Aspects of the Preservation of Electronic Publications. IViR Reports 7. University of Amsterdam, Institute for Information Law 1998 (www.ivir.nl/Publicaties/koelman/KBeng2.doc).
19 Jisc Model Licence for Journals (www.nesli.ac.uk/modellicence.html).
20 Archiving e-publications (www.jisc.ac.uk/index.cfm?name=project_epub_archiving).
21 OCLC/RLG Working Group on Preservation Metadata. Preservation Metadata and the OAIS Information Model: a Metadata Framework to Support the Preservation of Digital Objects. OCLC/RLG, 2002 (www.oclc.org/research/pmwg/pm_framework.pdf).
22 D. Martin. ONIX International Guidelines for Publishers Level 1. Book Industry Communication, 2001 (www.editeur.org/onixfiles1.2.1/ONIX%20Guide%20Level%201%20R1.2.1.PDF).

Adrienne Muir is a Lecturer in the Department of Information Science at Loughborough University (A.Muir@lboro.ac.uk). The information in this article was correct at the time of writing (end March 2003).