Journal of Law, Information and Science
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Law, Information and Science |
|
by DAVID LINDSAY
This article raises a number of points concerning the distinction to be made between electronic and non-electronic databases. The author considers that the differences are sufficient to create a separate regime for the protection of electronic databases. The writer then refers to a number of issues raised by the Copyright Law Review Committee on this matter. Two major aspects are discussed, the protection of works stored in electronic databases and the protection for authors of computerised databases.
Before commenting on some of the more specific issues, I would like to make a few general points about copyright and electronic databases. Firstly, if we were starting today with a completely blank slate, with no international treaties and no historical baggage, we would undoubtedly arrive at something very different from the current intellectual property regimes. Instead of copyright protection we might be more concerned with balancing rights to control of "information" given to information producers and information compilers; against rights of access to information given to competitors and end users. But we are not in that position. Because of this we find ourselves applying legal concepts and rules developed in a non-electronic age to an ever-growing variety of electronic media; and there is, naturally, not always a happy fit. Thus we have debates, for example, about whether information stored in an electronic or "non-sensate" form is in a material form, and so capable of protection.
The second general point I would like to make concerns the question of whether an electronic database is sufficiently different from databases in non-electronic form to deserve special treatment. The CLRC's draft report points to the existence of large card-based or file-based hard copy databases to suggest that electronic storage does not necessarily raise new or unprecedented copyright issues. The Committee suggests that the main distinguishing feature of a computer database is that it can be connected to a communications network, accessed via terminals and viewed without copyright necessarily being infringed. It would seem to me, however, that there are other, equally important, distinguishing features. For example, in relation to factual compilations, it is a basic principle that it is the selection or arrangement of facts, and not the facts themselves, that is protected. While the arrangement of facts in a non-electronic database is largely a matter of the author's choice, however, the arrangement of facts in an electronic database will often be dictated by the underlying logic of the software. In most cases, the value of an electronic database to users lies in whether it is comprehensive or complete, rather than in how it is arranged. It is the user who selects the relevant facts rather than the author. The skill and labour of the database compiler is in collecting comprehensive information, keeping it up to date, ensuring that it is accurate and entering it in a database that is made available to users; not the traditional skills of selection and arrangement exhibited by authors of manual databases. Another important distinction between manual and electronic databases, is that like all information in electronic form, electronic databases may be more easily pirated. On-line databases may be down-loaded and reproduced at low costs and high speeds over communications networks. Once a database is networked it may be expensive or impossible to detect acts of piracy. And, of course, there is the fact that the electronic databases may be continually, perhaps automatically, updated. Because of these distinctions I am firmly of the opinion that free of constraints imposed by international conventions, it would be preferable to establish a separate regime for the protection of electronic databases rather than protecting them in the same way as hard copy databases, as literary works. I believe that there are real dangers in attempts to graft the copyright protection of electronic media onto traditional schemes for protecting information in a written form.
The final general point I would make is that there is absolutely no point in establishing comprehensive laws for the protection of databases at a national level independent of international developments. In this sense the CLRC's draft report is prudent in not recommending radical changes to the existing laws. Effort should, instead, be focussed on finalising the long-overdue protocol to the Berne Convention for dealing with the problems of new information and communications technologies.
The following is a brief checklist of some more specific points raised in the CLRC's draft report and Anne's paper.
I do not think that anyone could disagree with the fact that there remain important uncertainties about the application of the current exclusive rights to works incorporated in a computer database. There is, at the outset, considerable uncertainty over the application of the reproduction right. On the one hand, even given some of the comments in the Autodesk decision,[1] the status of the requirement of "objective similarity" between a work and a copy, before there can be infringement, remains unclear. At a very basic level, therefore, it is unclear whether a work stored in electronic form can be a reproduction of a work in written form. On the other hand, as the CLRC point out, it is arguable that the reproduction right is extensive enough for a screen display to be a reproduction of a work stored in computer memory. Although the Committee recommended in favour of amending the Act to make it clear that screen displays are not a reproduction, it recommended against otherwise including a definition of "reproduction". While there are obvious dangers involved in attempting to clarify something as basic to copyright protection as the reproduction right, I feel that the addition of some new deeming provision in Section 21 of the Act, which deals with reproduction in different dimensions, should be looked at again; at least as a means for clarifying the conditions under which the electronic storage, or downloading, will constitute a reproduction.
There are also uncertainties about wether a screen display may be a public performance of a work stored in computer memory. Following the recent APRA decision,[2] the CLRC concluded that screen displays in a non-domestic context could be a public performance, and so recommended an amendment to make it clear that screen displays do not infringe the performance right in a work or database. This recommendation, which would differentiate between the screen display of works stored in computer memory and displays on television screens is not without its difficulties. Currently, it is an infringement of the performance right to display a work on a television screen in public. Why should works displayed on a television screen be protected while works displayed on a computer screen are not, particularly as computers may soon be incorporated in television sets? A possible reason for the Committee's approach to this issue was its hesitancy to confer a "use right" on owners of works included in databases. I would suggest, however, that a major problem with the performance right is the extremely generous interpretation applied by the courts to when a performance will be in public; and that differentiating between different kinds of display screens may be unnecessarily messy and technologically impossible. It should go without saying that any attempts to cut back existing rights should be carefully examined.
Finally, there are uncertainties about the scope of the diffusion right, with important consequences for on-line databases. At the outset, I must say that, given developments in delivery technologies, there is absolutely no justification in the kind of technology-based distinction currently made between the broadcasting right and the diffusion right. Delivery of information via fixed cable or optical fibre networks and over the air is increasingly interchangeable. The CLRC reached the conclusion that many databases would not fall within the definition of a diffusion service in section 26 of the Act, presumably on the basis that the transmission must be to the premises of "subscribers", and so does not apply within organisations or businesses. My reading of section 26, on the other hand, is that the diffusion right is not confined to subscription cable TV or radio services and probably applies to most commercial on-line database services, and, therefore, the need for clarification is this right deserves closer scrutiny than provided by the draft report.
The CLRC rejected a submission that reproduction of works by "electro-copying" be subject to a compulsory licence. Given the existing uncertainty about the rights of authors of works included in electronic databases, it is probably premature to establish a compulsory licence scheme. Nevertheless, I would suggest that there is now a substantial public interest in the collection of information in an electronic form, and in making the information available, such that any means for encouraging the development of electronic databases in Australia should be investigated. In this context, the CAL[3] for a statutory licence may deserve more detailed consideration.
Turning to the protection of electronic databases themselves, there are of course problems associated with the protection of all factual compilations, whether or not in electronic form. The problem arises because there can be no copyright in facts or information, only in the way in which the information is expressed or arranged. Under Anglo-Australian law, however, protection of an arrangement or selection of facts may sometimes result in protection of the facts themselves. The case law suggests that, rather than free riding on an author who has selected or arranged a factual compilation, anyone wishing to build on a factual compilation must go back to the source of the information and duplicate the work of the original author. This compares with the law in the US following the Feist telephone directory decisions, which rejected the "sweat of the brow" doctrine.
As the CLRC and Anne have pointed out, the standard for originality under Australian law is low by international standards. The central question in Australia is whether a compiler has imposed some order or coherence on a mass of raw data. There are however, many cases where mere lists or tables of information have been protected by copyright. Although it might be concluded that use of a computer program to compile a database might mean that the database is not original, as the author has expended insufficient effort, this is not necessarily the case. In Express Newspapers v Liverpool Daily Post,[4] for example, a set of computer numbers arranged in a grid form on a bingo card was protected, even though a computer had selected the numbers and arranged the grids. The computer was regarded as merely the tool of the operator. An issue that needs to be considered is whether the same database produced by manual means deserves greater protection than a database produced by a computer. Because computer-assisted compilation is undoubtedly more efficient than manual compilation, it would appear to be perverse to accord computer generated works less protection.
Another problem is the status of preparatory work. As the Ladbroke case[5] indicates, the courts may be reluctant to separate preparatory work from the work directly expended on the production of the database. Nevertheless, there may need to be a line drawn at some point in the chain of production of a database. For instance, should the author of a standard database program be regarded as a joint author of the database. Probably not; but under some circumstances, the skill and labour invested in producing database software may legitimately be taken into account in determining the originality of the database.
The CLRC concluded that, because of the extensive protection given to databases under current Australian law, there was no need to extend protection to databases lacking originality, in the manner suggested, for example, by the proposed "unfair extraction" right in the EC draft directive. The Committee, however, stated that it would consider the matter further in the light of comments received on the draft report. An unfair extraction right has a clear advantage over copyright protection in that it focuses on what is valuable to information producers, competitors and consumers alike: the content of the database rather than its form. It is also not restricted by the requirements of the Berne Convention, so may be much more flexible than copyright protection, for example, in connection with the duration of protection. There are, of course, constitutional difficulties involved in creating any comparable rights under Australian law. Given that the EC directive remains in draft form, and the needed to await developments on the Berne protocol, it would appear to be premature to establish an extraction right in Australia without further study and debate. It may well be that database producers are satisfied with the protection afforded by contracts with subscribers. In short, we need much more information about whether Australian database producers consider piracy to be a real and pressing problem, before thinking about establishing an entirely new regime.
The final major point I would like to mention concerns the protection of dynamic databases. Under Berne, of course, Australia is obliged to protect literary works for the life of the author plus fifty years. The CLRC formed the opinion that the duration of protection for dynamic databases would run from the time when the first edition of the database was made; and that additional information would only give rise to a new work if there was a substantial change in the database. The Committee added that the additional material might itself qualify for protection and abstained from making any specific recommendations about dynamic databases. The point I would make is obvious: as dynamic databases may be altered thorough gradual accretions rather than revolutionary changes, the point at which the changes may be substantial enough for there to be a new work is by no means clear cut. A legitimate question must be whether a database producer is to be penalised for supplying a database that is continually being upgraded in favour of a producer who issues quarterly or annual updates which may be substantial enough to make the database a new work, and so extend the duration of protection.
I agree with the CLRC that the problems of joint authors and ownership is not an issue confined to electronic databases. Ownership would appear to be an issue best left to contract law; and I would see no reason for introducing a special rule for commissioned works such as has been suggested for computer programs, at least without some further study of the effects of such a shift in the balance of power on the nature of the database industry.
[1] In their joint decision Mason C.J., Brennan and Deane J.J. said that "the test of originality is satisfied by the originality of the set of instructions and any unauthorised expression of it in language, code or notation will infringe copyright in the computer program". [1992] HCA 2; (1992) 22 IPR 163, p 164.
[2] Australian Performing Right Association Ltd. v Commonwealth Bank of Australia [1992] FCA 609; (1993) 25 IPR 157.
[3] Copyright Agency Limited.
[5] Ladbroke (Football) Ltd. v William Hill (Football) Ltd. [1964] 1 ALL ER 465.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlLawInfoSci/1993/20.html