A proposal for an open content licence for research paper (Pr)ePrints

Many academic papers that are to be submitted to refereed conferences and journals have been previously exposed to the author’s colleagues. The term ‘preprints’ has long been used to describe such documents. ‘Departmental Working Paper’ series were for many years a conventional vehicle for their publication. In the modern world, preprints are frequently transmitted electronically, variously as e–mail attachments and as files available for download via FTP or HTTP.

When a preprint is made available electronically, it is likely that the author provides the recipient not only with a copy, but also with a copyright licence. In most cases, however, the licence is only implicit, and the terms of the licence are unclear. This creates the potential for considerable uncertainties, and those uncertainties are of serious concern in the context of tension between for–profit publishers of refereed articles and the research communities that referee and edit them gratis, and depend on them for early access to information.

This paper briefly reviews the open content and ePrints movements, considers the interests of the various stakeholders, proposes a set of licence terms intended to satisfy the needs of all parties, and concludes that a particular Creative Commons licence type should be applied to all electronic preprints.

Contents
Introduction
Copyright, and its application to (Pr)ePrints
Requirements of a copyright licence for (Pr)ePrints
The licence specification
The process envisaged
Some further considerations
The applicability of existing licence sets
Conclusions

Introduction

It is vital to the progress of any discipline that papers documenting the outcomes of research be readily available to all interested parties. During the 1990s, conditions arose that threatened the achievement of that aim. Research burgeoned. Refereed journals greatly increased in number. Refereed journals became increasingly expensive. And libraries were forced by budget constraints to reduce their holdings.

Considerable concern arose about whether papers would continue to be accessible by other researchers, on a timely and continuing basis, and at an affordable price. A movement arose, commonly referred to as ‘Open Access’ and ‘ePrints’, associated with Harnad (1999). Its history and philosophy is well–documented in Suber (2004).

The Open Access movement involves the following elements:

proposals that authors electronically publish (or ‘archive’) preprints of their papers. In bygone years, this function was performed by Departmental Working Paper and Technical Paper series, which were printed, mailed, and made available at seminars. Electronic publication offers substantially greater reach and more rapid dissemination;

recommendations for the establishment of ePrints Archives (ePrints, 2002) by universities and other research institutions. Rather than tens of thousands of ephemeral, personal Web sites, the proposal is for hundreds of persistent, professionally managed, readily discoverable and interoperable locations (ePrints, 2003). This includes disciplinary repositories such as BioMed Central ( http://www.biomedcentral.com/), arXiv.org ( http://arxiv.org/) and CERN ( http://public.web.cern.ch/Public/Welcome.html), and institutional repositories such as DSpace at MIT ( https://dspace.mit.edu/index.jsp) and that at the Australian National University;
the publication of software that enables such ePrints archives to be managed;

recommendations for the use of the Open Archives Initiative (OAI, 2002) metadata standard, in order to support cross–discovery services;

recommendations that journal publishers sanction author self–archiving. This has already met with great success. See SHERPA (2002);

communications with governments, with a view to ensuring that government policy, and amendments to copyright law, support and not undermine open access to authors’ preprints.

In this paper, the term used for the ‘working paper’ is ‘(Pr)ePrint’. The reason for using an awkward compound is to ensure that the two key features are always in view: the fact that the work is intended as pre–publication of what it is hoped will later be a more refined document (a ‘preprint’); and, its electronic nature (an ‘eprint’).

The term ‘author’ is used in this paper to refer to the originator of the work. It is meant to be comprehensive, and to encompass all media types, including interactive media, for which other terms might be more appropriate. For the convenience of author and reader alike, it is used in the singular, and the question of multiple authorship is addressed only towards the end of the paper.

The term ‘publisher’ is used to refer to the person (natural or legal) that makes the work available to others. It is intended to encompass a wide variety of organisation types, operating under a wide variety of business models.

The term ‘consumer’ is used to refer to any person who makes use of the content of a (Pr)ePrint. A consumer’s use may involve actions that are not the subject of copyright (such as reading a copy of the work, and citing it, including creating a hyperlink to a copy of it), and actions that are subject to copyright law (such as reproducing it, or re–publishing it).

This paper commences by briefly reviewing copyright law as it relates to (Pr)ePrints. It then examines the interests of the various stakeholders, and proposes a set of requirements of a copyright licence for (Pr)ePrints. An outline description of the licence is provided, supported by a detailed specification. The role of the licence in the academic publishing process is presented. Some complexities are confronted. Finally, the suitability of available open content licence types is assessed.

Copyright, and its application to (Pr)ePrints

This section provides a very brief overview of the elements of copyright law, and explains its relevance to (Pr)ePrints. General guidance in relation to copyright and the Internet can be found in Dempsey (1996). More authoritative presentations of copyright law abound. Copyright is subject to international conventions, but in matters of detail it varies between countries. It is therefore not feasible to make authoritative statements that are precisely correct for every country. This paper treats copyright as though it were internationally consistent, and, as a result, some statements will be not quite correct and in the worst case even somewhat misleading, in relation to specific aspects of the law in some countries.

Copyright is a set of rights that vests in the originator of a work of a literary, dramatic, musical, artistic or cinematographic nature plus, in some countries, multimedia works; and, to what lawyers refer to as ‘subject matters other than works’: broadcasts and sound recordings. The key rights relevant to (Pr)ePrints are those to reproduce the work, to disseminate the work and to adapt the work.

Copyright ownership may be ‘assigned’ (i.e., gifted, bartered or sold) to someone else. Ownership may also be ‘gifted’, which places the object ‘in the public domain’.

A copyright owner may provide other parties with a licence to do specific things with a work. In particular, a licencee may be authorised (gratis, or for a fee) to make a designated number of copies, perhaps for a designated purpose, to translate it into another nominated language, or to re–publish it as part of a collection. Such licences are usually non–exclusive, by which is meant that the owner can grant licences to many parties, and may vary the terms of the licences depending on the circumstances.

The rights of a copyright owner are subject to a number of qualifications. Excerpts of a work may be able to be copied without a licence, provided that they are not ‘substantial’. Whether a copied part is substantial or not depends on qualitative factors as well as the size and the proportion of the work that is copied. Generally, such ‘fair use’ or ‘fair dealing’ provisions relate to copying and dissemination for such purposes as research, study, criticism or review.

Copyright infringement occurs when a particular action is taken, with respect to a ‘substantial part’ of a work, in the absence of legal authority to do so. The key actions that infringe copyright are reproduction (i.e. copying), dissemination (i.e. making available to others) and adaptation (including editing or translatinginto a new form, and substantively changing the content). Such infringements are subject to civil lawsuit.

The papers that are published in ePrints archives are, generally, copyrighted works. Later, if the paper is accepted for publication in a journal, the copyright in that paper, or in an adaptation of the (Pr)ePrint or a derivative work, may be assigned by the author to the journal’s publisher. There is a tension between the interests of the author and the publisher, and the achievement of appropriate resolutions to that tension lies at the heart of this paper.

It is likely that each person who accesses a (Pr)ePrint with the express or implied consent of the copyright-owner is granted a licence by the copyright owner. At the very least, the copyright owner is consenting to such copying as is necessary to enable the content to be displayed. Some countries’ laws permit this to be done without a licence; many do not. If a licence is required, then it can be argued that the consumer has an implied licence to produce a printed copy on paper, and perhaps also to disseminate copies to others. (An author who posts a (Pr)ePrint and takes no action to prevent such copying risks a court finding that this inaction constitutes permission).

The problem with current arrangements is that the terms of implied licences are not clear. Unless the licence terms are express, litigation would be needed in order to learn what the implied terms are. The outcomes are uncertain and could vary considerably between jurisdictions and even between cases in the same jurisdiction. In any case, litigation on such matters is highly undesirable, because of the costs, the delays, the adversarial relationships and the deflection of time and energy away from constructive work.

Generally, the authors of (Pr)ePrints actively want to permit certain uses of their works. This paper develops a particular form of copyright licence that reflects what authors want, but takes into account the other interests involved, in order to produce a realistic set of terms. It is intended that this licence be made available to researchers, and that it be promoted to the operators of ePrints archives.

Requirements of a copyright licence for (Pr)ePrints

Various categories of people and organisations have an interest in the terms of copyright licences for (Pr)ePrints. This section considers these various perspectives, in order to lay the foundation for a balanced licence specification. The categories of people and organisations considered are:

authors;

copyright owners;

publishers;

consumers; and,

the general public.

Authors

One expression of the situation is as follows:

"When copyright holders consent to [Open Access], what are they consenting to? Usually they consent in advance to the unrestricted reading, downloading, copying, sharing, storing, printing, searching, linking, and crawling of the full text of the work. Most authors choose to retain the right to block the distribution of mangled or misattributed copies. Some choose to block commercial re–use of the work. Essentially, these conditions block plagiarism, misrepresentation, and sometimes commercial re–use, and authorize all the uses required by legitimate scholarship, including those required by the technologies that facilitate online scholarly research." (Suber, 2004)

See also Willinsky (2002).

The licence needs to be devised so as to address the specific intentions of research authors relating to their (Pr)ePrints. The interests of the authors of ePrints can be interpreted as follows:

enable and encourage feedback on the content and presentation;

enable the filtering out of erroneous, imprecise, inadequate, uninteresting and low–value research;

enable the refinement and enhancement of potentially valuable research;

gain primacy for specific instances of discovery, invention and innovation;

gain citations in relevant literatures;

develop personal reputation; and,

contribute to organisational reputation.

The interests of authors in journal publication are more formal:

gain the recognition of one’s peers;

gain citations in relevant literature;

develop personal reputation;

enhance the likelihood of having research grant requests approved;

enhance recognition by potential employers; and,

contribute to organisational reputation.

Copyright owners

In many cases, the author will be the copyright owner, although some authors may (as I have done for many years) assign the copyright ownership to a company that they control. In some circumstances, however, the copyright owner may be the author’s employer, or some organisation with a contract with the author that stipulates that copyright in the author’s works are owned by that organisation.

Whereas the author’s perspective is intellectual and strategic, the copyright owner’s interests are financial:

enable the earning of revenue from uses of the work; and,

enable the earning of revenue from assignment of the copyright.

Publication of (Pr)ePrints assists these financial objectives only indirectly, in the senses of improving the quality of the work, pre–advertising its existence, and building up the author’s reputation. Meanwhile publication of a (Pr)ePrint may harm the financial objectives, especially if the final, refereed version is very little different from it, because a gratis resource is competing against a for–fee alternative.

The revenue objectives may be best served by not publishing through a journal at all, and instead making the information available selectively and for a considerably higher fee, e.g. by assigning the copyright, granting an exclusive licence to it, or making it available only in conjunction with services. The considerable tension that exists between open and closed research is not addressed further in this paper.

Publishers

Relatively small numbers of large works may be published as independent works, as is the case with doctoral theses. This paper focusses on the much larger numbers of article–length works. These are generally published in refereed journals, which appear periodically, typically monthly or quarterly. See Valauskas (1997).

The following categories of journal publishers can be delineated:

informal associations of a modest number of people with a common interest;

formally constituted not–for–profit associations of individuals, usually within a particular discipline or profession, which may have:

one significant publication;

multiple publications;

for–profit corporations, which may in turn be:

small businesses managing a small number of publications;

large, powerful corporations publishing substantial collections of journals.

For–profit publishing is not restricted to corporations. Some associations extract an excess from their publishing activities, which they use to cross–subsidise other activities. In a sample of 21 U.S. associations examined in Willinsky (2003), the accounts of 1999 or 2000 disclosed that six generated an excess in that year.

Considerable effort is involved in publishing a journal. The front–end of the production chain involves the receipt, management and processing of each submission, re–submission and correspondence; and the conduct and management of the assessment process. Each accepted paper requires production editing. Each issue requires production editing, production and distribution. Additional functions relevant to the journal as a whole include marketing, customer relationship management, archive management and governance. For more detailed analyses, see Odlyzko (1997) and Willinsky (2003).

These processes pre–suppose infrastructure in the form of an editor and editorial committee, a pool of referees, communications channels, norms for communications and formatting, production facilities, subscription list facilities, and one or more distribution mechanisms. They also depend on the original conception and articulation of the journal’s name, scope, philosophy and modus operandi, and preliminary negotiations within and beyond the intellectual community in order to bring the project to fruition.

It is feasible for journals to be produced by means of a mutual or co–operative, in a barter economy, and some significant e–journals have done just that, at least during their formative period. For example, the processes whereby First Monday is published have operated on that basis for a decade already. Moreover, open source software to support the publishing and management of journals is available, in many cases gratis, including Open Journal Systems (OJS, http://pkp.sfu.ca/ojs/) and others.

But many aspects of the infrastructure and processing are very wearing on volunteers, especially as the volume of submissions increases. Moreover, in a mutual operation, many costs are absorbed rather than dissolved. For example, production is dependent upon the use of equipment and software licences, variously for communications, production and distribution. eJournal economics, cost profiles, and business models are examined in Odlyzko (1997), Willinsky (2003) and Clarke (2005), and more generally in Clarke (2004) and Jeon–Slaughter, et al. (2005).

To cope with the diversity of approaches, this section separates the interests of publishers into three groups, reflecting the perspectives of mutuals, of not–for–profit associations, and of fully commercial operations. The interests are, by and large, cumulative, i.e. a for–profit publisher has additional interests, and these may result in some degree of compromise to the basic objectives, but without completely undermining them.

The interests of informal–group publishers can be inferred as follows:

serving the needs of authors by providing an appropriate outlet for their works;

serving the needs of individual consumers and the community by providing a venue in which relevant works can be found;

providing a nucleus whereby group–members can contribute expertise and effort on an ongoing basis in order to facilitate publication; and,

attracting sponsorship to defray costs.

The additional interests of more substantial association publishers can be inferred as follows

serving the needs of their members;

attracting subscription fees from members;

earning revenue from non–member subscriptions and per–access fees:
from libraries;

from consumers (individuals and organisations); and,

earning revenue from intermediary consolidators (who bundle large numbers of journals into collections, for which they charge subscription and/or per–access fees).

The additional interests of for–profit publishers can be inferred as follows:

earning revenue from the provision of a service, sufficient to more than offset the costs;

building the value of the corporate brand name by leveraging off the reputations of the journals under the corporation’s control;

earning super profits by owning critical titles, or a critical mass of titles in a particular area, such that some categories of consumer or intermediary simply have to pay what the publisher demands; and,

earning revenue from sale of the journal title, operations and archive.

The interests of association and for–profit publishers alike are harmed to the extent that accesses to a (Pr)ePrint substitute for accesses to the final version of the paper, and as a result revenue is reduced.

On the other hand, their interests are supported to the extent that the availability of the (Pr)ePrint pre–advertises the existence of the paper, and builds demand for the final version. There is also an indirect benefit whereby (Pr)ePrints enable the author to build and sustain a reputation (in much the same manner as ‘bloggers’ gain a following), and the journal achieves reflected prestige by virtue of the number of reputed authors who publish in it.

Consumers

The interests of consumers of academic writing can be interpreted as follows:

gaining access to relevant information and ideas;

gaining access to exemplars;

avoiding the delay and inconvenience involved in negotiating a copyright licence;

sustaining a sense of community; and,

gaining access to precise citations and references. During the current, transitional phase, the anachronism of citing page numbers continues. This may be replaced at some stage by a relative reference. In that case the published version will retain its importance, because an author who wishes to provide an accurate citation needs access to the authoritative version in order to generate the offset (Nelson, 1982; 1997).

Consumers have differential interests in (Pr)ePrints and final versions. The journal provides the authoritative, peer–approved version, whereas the (Pr)ePrint provides quick access to new information and emergent, as–yet–untested ideas, and a greater opportunity for interaction with them and their originators.

Generally, the more that the refereed version differs from the (Pr)ePrint, the greater the incentive for the reader to want access to the refereed version. Conversely, the less the difference, the greater the threat that the parallel availability of the (Pr)ePrint will reduce consumers’ need to gain access to the refereed version.

The general public

Finally, there is a more general public interest in an appropriate regime for ePrints. This has intellectual, social and economic dimensions. Innovation and progress are dependent upon discovery and invention generally, and on the accessibility of information about specific discoveries and inventions.

There is accordingly a general public interest both in (Pr)ePrints to ensure early exposure, and in reliable, financially healthy and long–lived journals that ensure the availability and archival of peer–approved works.

The licence specification

Based on the analysis in the previous section, this section expresses a design for a copyright licence that balances the many interests.

It is firstly necessary to appreciate the very wide range of possible open content licences. Clarke (2003) identifies the following dimensions of choice:

the power to grant licences;

protection of the integrity of the object;

control over reproduction;

control over dissemination;

control over adaptation;

balancing the encouragement of use against control over use; and,

liability management.

Very tight constraints on all of these dimensions represents an unequivocally ‘closed’ and ‘proprietary’ licence. Very loose constraints, on the other hand, represent a more or less ‘open’ licence; although holy wars are fought on precisely how far along each dimension the terms of a licence need to lie in order to be accorded the appellation ‘open’.

Although many possible licences could be generated by selecting different points on these dimensions, there is a counter–argument that people simply will not appreciate the nuances, and would prefer a simple choice. Indeed, attempts have been made to offer the simplest choice of all: ‘open’ versus ‘closed’.

This paper adopts the position that, in general, a requisite variety of licences is needed in order to reflect the diversity of categories of work and of circumstances. This is a similar conclusion to that previously reached by the Australian Educational Sharing Network (AEShareNet, http://www.aesharenet.com.au/) in 1998, and by Creative Commons ( http://creativecommons.org/) in 2001;

Further:

it would be highly desirable to provide a single licence for the specific purpose of (Pr)ePrints; but,

there may be a need for some flexibility, such as a small number of secondary alternatives for specific purposes, or the ability of the copyright owner to vary particular terms of the standard licence.

The Appendix defines the specific points on each dimension, effectively balancing various interests. Broadly, the choices are as follows:
the power to grant a licence is retained exclusively by the copyright owner;

the object may be copied in whole, but not in part; and each copy must carry the copyright notice with it, including the means of discovering the terms of the licence;

reproduction is permitted, but not for commercial purposes;

dissemination of copies is permitted, but not for commercial purposes;

adaptation is not permitted;

there are no limitations on who can be a licencee;

the licence is not time–limited, and is irrevocable;

the copyright owner provides no warranties or indemnities, other than any that may be imposed by law; and,

the licence is gratis.

Anyone who wants to do something that is precluded by this licence needs to contact the copyright owner and request specific permission. An author can grant multiple, specific licences to specific licencees, provided that care is taken to avoid conflict among them. (That is, if an author grants an exclusive licence, the author loses the right to grant licences to anyone else).

It is not intended that authors be required to make this form of licence available for their (Pr)ePrints. The copyright owner remains free to grant other kinds of licences either instead, or as well. (The copyright owner even remains free to leave the licence terms implicit, although, for the reasons explained earlier, this is against everyone’s interests).

What this paper argues is that this particular licence balances the various interests, that research communities and associations and operators of ePrint archives should encourage authors to use it, and that operators of ePrint archives should make the granting of such licences the over–ridable default for all (Pr)ePrints that they publish.

The process envisaged

The previous section defined the terms of the licence without reference to how it is to be used. This section describes the role of the copyright licence through the process of pre–publication, review and formal publication.

The author refines a paper to the point that it is ready for exposure and discussion. Generally, copyright subsists in the work whether or not a claim is made, and even whether or not the copyright owner appreciates that their creation has attracted some legal rights.

The author makes the paper available to others. This may be by providing copies directly to selected individuals, issuing an invitation to request copies from the author, publishing a discoverable copy on the author’s own Web site, or publishing a discoverable copy on some other site such as an ePrint archive.

Each person who acquires a copy with the express or implied consent of the author also acquires a copyright licence, and the ability to read the terms of the licence by following a link to a Web page.

The author may receive feedback, make presentations at seminars and conferences, and participate in discussions. Further, the author may reflect the insights gained by making modifications to the work and generating one or more revised versions.

The author submits the paper, but more likely a later version of the paper, to the editor of a refereed venue (a conference, a journal or a book).

There follow one or more rounds of refereeing, communications with the editor, and revision. These typically extend over some months. A significant proportion of such processes result in rejection, or in requirements for revision that the author never succeeds in satisfying. Hence a significant proportion of (pr)ePrints never give rise to a journal article. A significant proportion of those contain at least some discovery, invention, innovation, results or ideas that have value to some of their consumers.

The paper that was submitted for review, but most likely a yet later version of it, is accepted for publication in the refereed venue.

In preparation for publication, the publisher either:

requires the copyright owner to grant to the publisher a licence to the final version. The terms of the licence must at least enable the reproduction and dissemination of the work, and may grant much more substantial rights to the publisher. In particular, it may be an exclusive licence; or,

requires the copyright owner to assign the copyright to the publisher. In this case, the publisher may or may not grant a licence back to the author permitting some uses of the paper. Some associations, such as the Association for Computing Machinery ( ACM) and Institute of Electrical and Electronics Engineers ( IEEE), grant the author a quite liberal licence.

Irrespective of which approach the publisher adopts, the copyright arrangements in respect of the final version of the article do not affect the (possibly many) existing licences relating to the (Pr)ePrint. Nor do they affect the ongoing availability of the (Pr)ePrint and of licences in relation to it. They may, however, preclude the provision of later versions of the work, and in particular of the version that is to appear in the refereed venue.

When the refereed article is published (perhaps after a considerable delay), the refereed article may include reference to the (Pr)ePrint (e.g. in the Acknowledgements section), possibly including a hyperlink.

The author then adds to the (Pr)ePrint a citation to the refereed article. It is beneficial if this includes a link. The refereed article may be openly available or it may require a subscription or other payment.

Some further considerations

This section considers several factors that were assumed away in the earlier part of this paper, in order to simplify the presentation.

The first issue concerns copyright ownership. This may rest with the sole author, or if there are multiple authors then ‘jointly and severally’ among them. Alternatively, the terms of employment of any of the authors may be such that their employer owns the copyright from the time the work comes into existence. A further possibility is that the author(s) may be bound by the terms of contract or by contract law to assign the copyright in the work to some other party. This would typically be the commissioner of the work, or the provider of the funding necessary to enable its development. This paper has avoided the complexities of authorship and copyright ownership, and has assumed that there is a sole copyright owner, or at least a sole human acting as agent for all copyright owners.

The second complexity arises from the way in which copyright law treats revisions to a work. When modifications are made, they may give rise to a new version of the same work; or, they may give rise to a new work in which a new set of rights exists. The threshold for the emergence of a new work is unclear in law, and can only be determined by the courts. The law therefore unfortunately leaves scope for wasteful arguments as to whether the journal article is a separate work from the (Pr)ePrint, or merely a later version of the same work.

The third consideration is that publishers might force authors to withdraw (Pr)ePrints as a condition of publication in the refereed venue. Possibilities include:

at the back end of the process, the publisher’s standard–form copyright assignment or licence might include such a term. This would be harmful to research, because it would result in the withdrawal of early versions of works that have been published (or, even worse, that are at some stage going to be published); and,

at the front end of the process, the submission of papers for consideration by the editor and referees might include a term precluding pre–publication in the form of a (Pr)ePrint. This would represent an even more serious constraint on the accessibility of research outcomes, because it would actively discourage the lodging of works in ePrints archives, not only those that eventually achieve the status of a refereed article but also those many that do not.

Relatively few publishers currently appear to impose such terms, and only a few have such high brand value or market dominance that they could mobilise sufficient market power to do so. Authors might of course voluntarily suppress their work, by failing to make a (Pr)ePrint available, at least in part because of nervousness about what the editor or publisher of the targeted journal might think. On the other hand, awareness of the importance of openness is being strongly reinforced by the efforts of the Open Access and ePrints movements, and the surge of publicity in 2003–2004 surrounding the Creative Commons initiative.

The applicability of existing licence sets

It is possible that a currently available open content licence may fit the need. Leading examples discussed here are a specifically Australian set called AEShareNet, and the originally U.S. but now multi–national Creative Commons set.

AEShareNet

The Australian Educational Sharing Network is a collaborative initiative among the nine Australian governments. A set of open content ‘licence protocols’ was designed in 1998, and has been operational since 2002. Most are convenient ‘instant licences’. The company’s Web site provides an overview of the complete set of licence types (AEShareNet, 2002).

One of particular interest from the viewpoint of this paper is the Free for Education (FfE) licence. This licence is very similar to the specification developed in this paper, but differs in the following ways:

it does not authorise the licencee to disseminate copies on to third parties. This is a constraint that is expressly discussed in the specification, but it is probably a second–best alternative;

it permits minor editorial changes (but not enhancements). This is a practical matter rather than one of principle, and can be seen as an enhancement to the specification; and,

it authorises copying and use only within the context of "a structured program of learning and/or teaching for the benefit of a learner" (including "structured and purposeful self–education"). This is in apparent conflict with the specification developed in this paper, because when researchers use other researchers’ papers they do not generally consider themselves to be pursuing an educational activity.

It is feasible to overlay further terms over the standard FfE licence in order to overcome these limitations. For example, the licence offered in the heading of this paper is sufficient to address the third of the differences noted above.

Alternatively, AEShareNet could be approached to create a Free for Research (FfR), or a Free for Education and Research (FfER) licence, whose terms match the specification in the Appendix to this paper.

Creative Commons (CC)

The Creative Commons initiative has attracted a great deal of attention since 2003. It now provides U.S. licences and licences subject to the laws of currently 14 other countries, including Australia. Its Web site, at http://creativecommons.org/, provides an overview of the terms of the various licences (CC, 2001).
There are many variants. Some are related to the medium in which the work is expressed. Others reflect different choices available to the copyright owner concerning the licencee’s rights to adapt the work, and to use it for commercial purposes.

The closest to the specification developed in this paper is the Attribution/NonCommercial/No Derivative Rights (By–NC–ND) variant. The Web site provides for that licence type both an overview (CC, 2003a) and the full terms (CC, 2003b). This paper is available under the equivalent Australian licence, for which see the overview (CC, 2004a) and the full terms (CC, 2004b).

This licence differs from the specification in the Appendix to this paper in the following way:

it permits any modifications technically necessary in order to reproduce the work in other media and formats. This is a practical matter rather than one of principle, and can be seen as an enhancement to the specification.

It appears that this licence type, in any of the U.S. and other national versions, satisfies the specification in the Appendix to this paper.

Conclusions

When authors make (Pr)ePrints available, they generally grant a copyright licence to the recipients, whether they realise it or not. It is important that the terms of that licence be clear, and that the terms balance the interests of the various parties that have a stake in the undertaking of research publication.

This paper has analysed the requirements of such a licence, and developed a specification. It has considered available alternatives, and concluded that the Creative Commons Attribution/NonCommercial/No Derivative Rights (By–NC–ND) variant is appropriate to the purpose.

This proposal needs to be exposed, in journals, appropriate conferences, submissions to the governing bodies of individual ePrints Archives, and submissions to relevant associations.

Consideration needs to be given as to whether the standard ‘Creative Commons – Some Rights Reserved’ title, icon and text should be used, or whether instead negotiations should be entered into with the Creative Commons organisation with a view to establishing a purpose–specific title, icon and text, such as the ‘Creative Commons ePrint Licence’.

About the author

Roger Clarke is Principal of Xamax Consultancy Pty Ltd, Canberra. He is also a Visiting Professor in the E–Commerce Programme at the University of Hong Kong, Visiting Professor in the Baker & McKenzie Cyberspace Law & Policy Centre at the University of New South Wales, and Visiting Fellow in the Department of Computer Science at the Australian National University (ANU).

Since Xamax’s formation in 2000, the author has been Chair of the Ministerial company that implemented the set of Australian Educational Sharing licences, AEShareNet Limited.

Acknowledgements

An earlier version of this paper was presented to the ANU Department of Computer Science on 29 September 2004. The paper has benefited significantly from comments provided in response to the (Pr)ePrint by a wide range of colleagues active in the Open Content, ePrint, AEShareNet and Creative Commons communities, and from referees’ comments.

References

AEShareNet, 2002. "Short Licence Comparison Table," at http://www.aesharenet.com.au/coreBusiness/whatWeDo/027comparison.asp, accessed 26 March 2005.
Creative Commons (CC), 2004a. "Commons Deed – Attribution–NonCommercial–NoDerivs 2.0 Australia," at http://creativecommons.org/licenses/by-nc-nd/2.0/au/, accessed 26 March 2005.
Creative Commons (CC), 2004b. "Legal Code – Attribution–NonCommercial–NoDerivs 2.0 Australia," at http://creativecommons.org/licenses/by-nc-nd/2.0/au/legalcode, accessed 26 March 2005.
Creative Commons (CC), 2003a. "Commons Deed – Attribution–NonCommercial–NoDerivs 2.0," at http://creativecommons.org/licenses/by-nc-nd/2.0/, accessed 26 March 2005.
Creative Commons (CC), 2003b. "Legal Code – Attribution–NonCommercial–NoDerivs 2.0," at http://creativecommons.org/licenses/by-nc-nd/2.0/legalcode, accessed 26 March 2005.
Creative Commons (CC), 2001. "Licenses Explained," at http://creativecommons.org/about/licenses/fullrights, accessed 26 March 2005.
R. Clarke, 2005. "The Cost–Profiles of Alternative Approaches to Journal–Publishing," Fiesole Collection Retreat Series, number 7, Melbourne, 29 April 2005, at http://www.anu.edu.au/people/Roger.Clarke/EC/JP-CP.html, accessed 15 April 2005.
R. Clarke, 2004. "Open Source Software and Open Content As Models for eBusiness," Proceedings, 17th International eCommerce Conference (Bled, Slovenia, 21–23 June), at http://www.anu.edu.au/people/Roger.Clarke/EC/Bled04.html, accessed 26 March 2005.
R. Clarke, 2003. "Copyright: The Spectrum of Content Licensing," at http://www.anu.edu.au/people/Roger.Clarke/EC/CCLic.html, accessed 26 March 2005.
G. Dempsey, 1996. "A Copyright Guide for Internet Users," at http://www.anu.edu.au/people/Roger.Clarke/II/Dempsey-copyright.html, accessed 26 March 2005.
ePrints, 2003. "Institutional Archives Registry — Home," at http://archives.eprints.org/, accessed 26 March 2005.

ePrints, 2002. "ePrints Home Page," at http://www.eprints.org/, accessed 26 March 2005

S. Harnad, 1999. "Free at Last: The Future of Peer–Reviewed Journals," D–Lib Magazine, volume 5, number 12 (December), at http://cogprints.org/1685/00/12harnad.html, accessed 26 March 2005.

H. Jeon–Slaughter, A.C. Herkovic and M.A. Keller, 2005. "Economics of scientific and biomedical journals: Where do scholars stand in the debate of online journal pricing and site license ownership between libraries and publishers?" First Monday, volume 10, number 3 (March), at http://www.firstmonday.org/issues/issue10_3/jeon/, accessed 26 March 2005.

T.H. Nelson, 1997. "Transcopyright: Dealing with the Dilemma of Digital Copyright," Educom Review, volume 32, number 1 (January/February), pp. 32–35, and at http://www.educause.edu/pub/er/review/reviewArticles/32132.html, accessed 26 March 2005.
T.H. Nelson, 1982. Literary Machines. Sausalito, Calif.: Mindful Press.

A. Odlyzko, 1997. "The economics of electronic journals," First Monday, volume 2, number 8 (August), at http://firstmonday.org/issues/issue2_8/odlyzko/, accessed 26 March 2005.

Open Archives Initiative (OAI), 2002. "Open Archives Initiative," at http://www.openarchives.org/, accessed 26 March 2005.

SHERPA, 2002. "Publisher copyright policies & self–archiving," at http://www.sherpa.ac.uk/romeo.php?all=yes, accessed 26 March 2005.

P. Suber, 2004. "Open Access Overview," at http://www.earlham.edu/~peters/fos/overview.htm, accessed 26 March 2005.
E.J. Valauskas, 1997. "Waiting for Thomas Kuhn: First Monday and the evolution of electronic journals," First Monday, volume 2, number 12 (December), at http://www.firstmonday.org/issues/issue2_12/valauskas/, accessed 26 March 2005.
J. Willinsky, 2003. "Scholarly Associations and the Economic Viability of Open Access Publishing," Journal of Digital Information, volume 4, number 2, at http://jodi.ecs.soton.ac.uk/Articles/v04/i02/Willinsky/, accessed 26 March 2005.
J. Willinsky, 2002. "Copyright Contradictions in Scholarly Publishing," First Monday, volume 7, number 11 (November), at http://www.firstmonday.org/issues/issue7_11/willinsky/, accessed 26 March 2005

Appendix: Draft Specification for an Open Content Licence for Research Paper (Pr)ePrints

This licence is granted when a person acquires a copy of the relevant (Pr)ePrint with the express or implied consent of the copyright owner.

1. Control of the Work

Non-Exclusive
By granting a licence, the copyright owner does not lose the right to grant licences to other people, i.e. the copyright licencee acquires a non–exclusive licence.

No Sub–Licencing
The licencee is not granted any power to issue copyright licences in the work, i.e. if a licencee disseminates a copy of the (Pr)ePrint to another person, the third party does not acquire a licence from the second party, but automatically acquires a licence from the copyright owner.

2. Integrity Protection

Copy in Entirety
All reproductions made by the licencee must be of the complete (Pr)ePrint, not part of it. The purpose of this term is to ensure that the reader does not access the information out of its intended context.

This does not affect rights under the fair use and fair dealing provisions of copyright laws, which enable reproduction of the paper, or in most cases of small segments of the paper, for various purposes. Those rights exist under law, and cannot be removed or varied by a licence.

Copyright Notice
Each reproduction made by the licencee must carry the copyright notice with it. The purpose of this term is to ensure that the terms of the licence, and the constraints on its use, are readily discoverable by everyone who acquires a copy.

3. Reproduction Control

Permission to Copy
The licence grants permission to the licencee to make copies of the work. This is necessary for the purposes of download, display and printing. It extends to the making of further copies. These may be for personal use (e.g. a printout of a downloaded copy to read on a train or plane, or a further soft–copy to put on another workstation used by the same person). Alternatively, the copies may be for dissemination to other parties, but this is subject to the conditions explained below.

The licence is silent on such questions as backup copies, caching, proxy–servers, search–engines, and other network paraphernalia (because those are broader issues that are appropriately addressed by copyright law as a whole rather than by individual licences).

Formats and Media
The licence grants permission to the licencee to make copies of the work in any format, on any medium.

No Commercial Use
The licence precludes the use of the copy for any commercial purpose. It permits only personal and not–for–profit use. For clarity, copying and reading by an individual who is employed by a for–profit organisation is not to be interpreted as commercial use, and is permitted.

4. Dissemination Control

The licence permits dissemination of a copy only if all of the following conditions are satisfied. Dissemination encompasses inclusion in a compilation such as a book of readings:

the dissemination must not be undertaken for any commercial purpose. Dissemination as part of a ‘reading–brick’ or electronic reserve for students is permissible, provided that any fee that is charged is of the nature of cost–recovery;

the whole paper must be disseminated, not part of it. (This licence term does not, and cannot, affect those forms of dissemination that are permitted under fair use or fair dealing provisions);

the copyright notice must be carried with the copy; and,

the dissemination must not purport to offer a copyright licence to recipients, but rather each recipient acquires, and is bound by, this licence.

[Comment: An alternative would be to preclude dissemination in any manner, and force all readers back to the original source. This is impractical, and may be excessive in any case. The alternative approach adopted here ensures that all readers are subject to this licence, irrespective of how they acquire the copy.]

5. No Adaptation

The licence precludes adaptation of any kind. This means that it is not permissible to even edit the work (e.g. for corrections of apparent errors), to annotate it, or to translate it into another language. If an individual or organisation wishes to add value to the work, then they need to approach the copyright owner for a specific–purpose licence.

The purpose of this term is to ensure that the author retains control, and can choose which individuals and organisations they collaborate with and publish through. This also reserves for the author the possibility of ‘going proprietary’ with an enhanced or extended version of the work, e.g. by selling the copyright and/or associated expertise to a high bidder, while sustaining open accessibility to the original (Pr)ePrint.

6. Usage Limitations

Availability of the Licence
The availability of the licence is not restricted in any manner. For example it is available to people and organisations in any geographical location and any legal jurisdiction, irrespective of the category of person or organisation, irrespective of the field of endeavour in which the person or organisation works, and irrespective of their purposes in acquiring the licence (subject to the preclusion on reproduction and dissemination for commercial purposes).

Term and Revocability
The licence is not subject to any time–limitation.

The licence is non–revocable. In particular, the licence survives assignment to another party of the copyright in this work, in any subsequent work, and in any derivative work.

These features are important, because without them the publisher of a refereed version of the work might be able to impose a legal constraint on flows of information within the research community.

A licence that is in the form of a contract might be ineffective in achieving non–revocability. (This is because a contract requires consideration, and the recipient might well be construed not to have offered any). The licence may therefore need to be expressed in the form of a deed rather than a contract (because a deed is valid notwithstanding the lack of consideration being offered).

7. Liability Management

The copyright owner provides no warranties or indemnities to the licencee, other than any that may be imposed by law.

8. Pricing

The licence is granted gratis, free of both one–time fees and repetitive fees of any kind.

This does not preclude the author charging a fee for specific–purpose licences, or for licences for a later, or enhanced, or extended version.

Editorial history

Paper received 15 April 2005; revised 1 May 2005; accepted 14 May 2005.
HTML markup: Susan Bochenski and Edward J. Valauskas; Editor: Edward J. Valauskas.

This work is licensed under a Creative Commons License.

A proposal for an open content licence for research paper (Pr)ePrints by Roger Clarke
First Monday, volume 10, number 8 (August 2005),
URL: http://firstmonday.org/issues/issue10_8/clarke/index.html