Part of the basic philosophy of the eLib programme is to explore issues to do with copyright, to engender culture change amongst librarians, users and copyright owners about copyright, and to explore novel contractual or technical means to overcome the difficulties that copyright poses. In this article, I look at some of the major project lines and the copyright issues that might be raised from them.
The major issue that has arisen is negotiating rights from publishers for scanning the material in that subsequently is offered under on demand publishing. Publishers have shown remarkably wide variation in their reactions to requests for permission, even when it is pointed out this is for a short time scale for experimental purposes only. Amongst the problems raised by publishers are:
Some publishers are also insisting that at the end of the experiment all material that has been converted into machine readable form be destroyed, something that could be addressed by having some sort of escrow arrangement, whereby instead the material is deposited in some repository to which no one can have access without the copyright owner's permission.
Even more surprising, perhaps, is that a given publisher may react in entirely different ways to two different approaches from two different On Demand project teams.
One issue that has arisen recently in connection with this area, but applies equally to other project lines, is the role of the creators. We think of publishers owning copyright, so they are the ones to talk to. Often an author or artist also has rights - they may not have assigned all their copyright rights to a publisher, and in any case, will almost certainly retain Moral Rights themselves. In recent weeks both the ALCS and DACS, who represent authors and artists respectively, have approached FIGIT asking that their members' views are taken into account.
The only way to ensure that you do not get into any problems about authors or artists is to INSIST and I use the word advisedly - on a clause in any On Demand contract as shown in my first Handout:
The Publisher represents and warrants to X that it is the owner of the copyright in the Information or that it is duly licensed to use the copyright material contained in the Information and that the Information used as contemplated in this Agreement does not infringe any copyright or other proprietary or intellectual property rights of any natural or legal person.
The Publisher shall indemnify and hold X harmless from and against any loss, damage, cost, liability or expense (including reasonable legal and professional fees) arising out of any actual or alleged infringement of such rights. X shall promptly inform the Publisher of any such infringement or suspected or threatened infringement upon X becoming aware of the same.
This indemnity shall survive the termination of this Agreement for any reason. This indemnity shall not apply if X has amended the Information in any way.
If the publisher refuses such a clause, or something similar, don't sign a deal with them. I cannot put it simpler than that.
Here the main issue is copyright assignment by the authors. As you may be aware, there is considerable debate right now whether authors should assign copyright to journal publishers at all, and if so, whether they should confine themselves to assigning just very limited rights. FIGIT does not wish to be prescriptive; we want to encourage all sorts of different models; however we would be disappointed if all the electronic journal projects adopted the current publisher stance of; you must assign all copyright to us if you want this item published by us.
Let me make one thing clear - if an author submits an article to an electronic (or any other type of) journal, and there is no clear contract stating that copyright is being assigned, then in copyright law, the author retains copyright. Therefore, in theory the electronic journal is infringing copyright by publishing the article! There isn't necessarily any implied transfer of copyright just because someone has submitted an article to you. Therefore, the one thing you cannot do is ignore this issue. You MUST have some policy in place.
Besides having some clause making it clear that you as the e journal publisher have some rights to reproduce the material on your electronic journal, you will need some standard contractual clauses with your authors along the following lines - and remember, this time you are the publisher:
The author warrants to the Publisher that it is the owner of the copyright in the material submitted, and that the material does not infringe any copyright or other proprietary or intellectual property rights of any natural or legal person. The author indemnifies and holds the Publisher harmless from and against any loss, damage, cost, liability or expense (including reasonable legal and professional fees) arising out of any actual or alleged infringement of such rights save where this is the direct result of any amendment of the material done by the Publisher without the agreement of the author.
The author further warrants to the Publisher that publication of the material will not contravene any laws, including but not limited to the laws of libel, defamation and contempt of court (or concepts approximating thereto). The author shall indemnify and hold the Publisher harmless from and against any loss, damage, cost, liability or expense (including reasonable legal and professional fees) arising out of any illegality or alleged illegality save where this is the direct result of any amendment of the material done by Publisher without the agreement of the author.
This really is to protect your own interests; I strongly commend these clauses to you.
This is potentially a copyright minefield. Librarians have a unique position under the Copyright Act. Unlike the rest of humanity, we have a guaranteed immunity against being sued for infringement by copyright owners, as long as we follow certain boring bureaucratic procedures concerning copying for our patrons. We are explicitly allowed to make single copies - which can, in theory, either be a hard copy or an electronic copy incidentally, but it must be a SINGLE copy - but only on the following conditions:
Note that requests on the phone, by e mail, etc. are unacceptable. The form must be personally signed by the individual. This means a secretary's signature is not acceptable. It also means a stamped signature is not acceptable. Whilst in theory the form should be received first, then the copy made, in practice the library can make the copy first, as long as it is exchanged with a signed form from the patron. The library should store these signed forms for six years, and be prepared for them to be checked.
What does this mean in practice? It means that whilst in theory you can create a copy electronically and deliver it electronically, or in printed form, in practice to avoid the risk of making multiple copies - which would be infringement - you must control the creation and distribution of electronic copies very carefully indeed. If you plan to go down this route, I suggest you consult with your own lawyers, or with me and Chris Rusbridge before you proceed. The law is totally explicit on the need for a personally signed form, so there is no way to get round this, incidentally.
The main issue here is that you will have invested time and effort in creating certain types of metadata - some indexes or value added catalogue records in a particular subject area. The products of these efforts, under copyright law, are your copyright, but things get more complex when there are several partners, each of whom has contributed to the resulting network resource tool. You should sign an agreement amongst yourselves as to the ownership of the copyright in what you create; I commend the TLTP Copyright Guidelines, which can be obtained from TLTP at HEFCE, as a good model for such agreements; please don't ignore this issue, as should your network tool turn out to be another Yahoo!, you will surely want a clear agreement so you can divvy up the money when you float your company on the Stock Exchange. This may sound flippant, but it isn't. I have no idea if the developers of Yahoo! had such an agreement in place, but to avoid problems, get one set up now.
The only other thing to say is that you should be vigilant in protecting your metadata and value added information against infringement, if you believe you have created something of value that you wish to protect for financial reasons, or to protect your own reputation.
Incidentally, exactly the same remarks apply to those projects in the education and training action line if you are creating products that will assist in education and training, such as multimedia packages, and you anticipate they will be sold, or widely used and copied. I would add that the copyright situation of multimedia is FAR more complex than simple Literary Works - by which I mean texts and software. I suggest if you are creating multimedia products and are concerned about the copyright situation, check the TLTP Guidelines and chat to me, as I do not have time to discuss the issues here.
Let me finish off by briefly mentioning ECMS - Electronic Copyright Management Systems. Publishers have difficulty in agreeing to any license for the distribution in electronic form material for which they hold the copyright. There are two major reasons stated for their concern. The first is the worry that the material will be copied and/or re-disseminated in an unauthorised manner, and therefore, by implication, the publishers will lose sales.
The second concern is that material will be amended and will then be passed off as new material, and it will be difficult to demonstrate that the material had originated from that material that the publisher owned rights to.
An ECMS addresses these issues. One type is software incorporated into word processing, computer typesetting and DTP softwares, as well as Document Image Processing equipment. This type of ECMS would automatically tag the document in a tamper proof fashion. This could be read by anyone to identify the original author and/or copyright owner of the material, and to identify who had made any amendments to the document. An audit trail would thus be clearly identified. Another type of ECMS is software used solely to govern or control distribution of the work, which may be in printed or electronic form.
The systems will also provide copyright management information to inform the user about authorship, copyright ownership, date of creation or last modification, and terms and conditions of authorised uses. Once information such as this is affiliated with a particular work, users will be able to easily address questions over licensing and use of the work.
No well established Electronic Copyright Management System currently exists, but there is a lot of research going on in the field in the USA, Europe (mainly funded by DGXIII) and in the UK (funded by eLib). eLib is funding two separate projects - a review of the whole area, carried out by Bill Tuck that will be completed in the next few weeks, and a medium term project called ERCOMS, based in my University and starting in a few weeks' time, for the development of ECMS for a particular application - electronic reserve systems.
There is no doubt that the copyright owners are keen on the idea of ECMS. In the USA, the Digital Rights Management group, representing major players such as UMI, West, Reed, Lexis-Nexis, Dun and Bradstreet and McGraw Hill are actively developing policies and standards in the field. It is a moot point whether technological fixes such as ECMS will ultimately prove viable, but it is important that those involved in eLib and similar projects keep a close eye on developments in this field.
Below is a reply to comments made by Nick Joint about this article.
Nick's original response is tabbed with >'s - Charles response to Nicks response appears as normally tabbed text
> May I congratulate both the author and the journal for publishing such
> an informative and helpful article on the thorny topic of electronic
> copyright in the most recent issue of ARIADNE (May 1996). Could I also
> tease out some points from the section outlining the statement
> to be included in On Demand Publishing contracts with publishers, which
> I copy (for the purposes of fair comment and scholarly criticism!)
> Firstly, it would be interesting to know how to fill in a variable for
> the value "X" in the contract statement! Elib-funded projects are
> funded on an avowedly collaborative basis, which gives rise
> to problems over liability- just who is responsible for what in the
> consortium model, and how do you regulate such responsibility? If a
> lead site handles the drawing up of contracts, should that site be the
> elusive "X", or should the whole consortium be named as the potentially
> offending "X" in question? Even if the consortium is named as the entity
> with whom the publisher makes the agreement, should a separate document
> regulating responsibility be looked to as the main means of regulating
> liability between the members of the consortium, and to what extent
> should the distribution of the liability between members of the
> consortium be explicitly stated in the publishers' agreement (since it
> would be vital for the publisher to know just who is responsible for
> what in the contract they are signing)?
The contract has to be a natural or legal person. A consortium team may have no formal status (e.g. as a partnership or registered companhy), so I would always recommend that the contract should be signed with the lead institution. You might then want to have letters of agreement between that lead institution and the partner institutions outlining their share of their responsibilities (this is how SCOPE, for example, has done it).
> Secondly, the phrase limiting the danger of liability through
> infringement of moral rights ("the Information used as contemplated
> in this Agreement does not infringe any copyright or OTHER
> PROPRIETARY OR INTELLECTUAL PROPERTY RIGHTS of any natural or legal
> person...") makes it very important that the exact nature of the use
> of the information is explicitly laid out somewhere in the agreement.
> Since many On Demand Publishing Elib projects will be taking excerpts
> rather than complete works of authors, it is important that
> publishers/authors be aware that they are being asked for permission for
> partial copying of a work, and for a statement that moral rights (rather
> than "copying rights") are not being infringed in any way. This is
> especially true in view of the final sentence of the clause, which
> renders the protection given by the clause invalid if the work of the
> author is amended in any way- partial copying may be taken as a
> form of amendation to the work, and thus prohibited by the terms
> of the clause itself.
Yes, absolutely right, and the question of Moral Rights should have
been adressed at far greater length in my article. The digitising
may well be permitted by the copyright owner, but that does not stop
Moral Rights arising if only a selection of the book is
scanned in. The author may object on the grounds that this is
derogatory treatment. How to deal with this? *Either* write to the
author in person outlining what you plan to do and ask for their
agreement to it; *or* go ahead anyway and hope that what you are
doing is not derogatory treatment. the latter is risky! On the plus
side, Moral Rights only apply if the person has asserted them. By
checking the book to see if the author has asserted his/her Moral
Rights. if there is no mention of such an assertion, you can ignore
the issue of Moral Rights.
> Lastly, the clause is an excerpt from a contract between
> an Elib project and a publisher. Because (as the article
> specifically states) authors can assign copyright but
> reserve moral rights, for a publisher to sign the contract
> in legal safety, they would have to contact their authors to gain
> assurance that (for example) partial copying of their work does
> not infringe their moral rights. Only then could the publisher sign
> the contract. Isn't this rather a lot of work for them? Surely there
> is a danger of them walking away from the amount of administration
> involved in this sort of agreement?
Publishers are completely free to assign copyright as they see fit.
They are not going to get involved with the arguments on Moral Rights
- that is between the eLib consortium and the author direct - so it
does not cause a problem for the publishers, other than perhaps to
supply the consortium with the publisher's address, or act as a post
box for communications.
> I make these points having read recently that the conductor Claudio
> Abbado is suing his record company for "publishing" a cd of excerpts
> from Abbado's cycle of Mahler symphonies. These excerpts
> were compiled allegedly in breach of his moral rights. If only they
> had contacted Charles Oppenheim before producing the album...
Yeah, and if only I charged a proper consultancy fee... :-<