Lapas attēli
PDF
ePub

(1) Under what conditions should the input of copyrighted documents into a computer system be deemed to infringe the copyright?

(2) Under what conditions should the output of such documents
or portions of them from the computer system be deemed
to infringe the copyright.

(3) Where permission from the copyright owner is required for the use of a document in a computer system, should such permission be obtained before input, or should it suffice to obtain permission before output?

A.4.5.2 Input and Output of Documents as Infringement. It will be perceived that, in the main, the questions concerning the input and output of copyrighted documents are substantially the same as those pertaining to the input and output of copyrighted data bases. In fact, data bases are a category of complete documents in themselves. Accordingly, the discussion of these questions above in relation to data bases would be applicable to the storage and retrieval of the full text of copyrighted documents in computer systems. As to input, see sections A.4.4.2.1 and A.4.4.3.1 through A.4.4.3.4. As to output, see sections A.4.4.4.1 and A.4.4.5.

One difference, however, may be noted. Whereas the output from a data base will usually consist of a few only of the mass of items in the copyrighted compilation of data, the output in the case of a document will ordinarily be of the entire work. In the latter case there would be no question of fair use. However, the user of a computer system could not be charged with infringement for his extraction from it of a complete copy of a copyrighted document as long as the system is authorized to provide its users with such documents. But if he then used the copy so extracted to make further copies of the document, he would thereby be infringing the copyright. And if a person not entitled to use the system did so surreptitiously to produce copies of copyrighted documents, he would be committing an infringement of the copyright as well as an offense against the system itself. It seems likely, however, that wrongful acts of this nature would often escape detection. (Cf. section A.4.4.4.2.)

One more point is in order here. We suggest that a publisher would be well advised, when he licenses the input and output of copyrighted documents in a computer system, to require the system to have its computer programmed to reproduce the copyright notice on each reproduction of the work as output. (Cf. section A.4.4.1.2.)

A.4.5.3 Input or Output as Occasion for Obtaining License. We have adverted earlier to the discussion, in the 1967 Senate hearings on the copyright revision bill and elsewhere, of the question whether the input of a copyrighted document into a computer system should require a license from the publisher, or whether input should be free though a license will be required for output. The arguments advanced for free input, enunciated mainly by members of the academic community, may be summarized as follows:

(1) Works may be put into computers for the purpose of a
noninfringing manipulation of the work within the com-
puter that will not result in any output of the work
itself. Known examples include the analysis of the
text of a work to show the characteristics of an author's
style or the frequency of word uses, or the preparation
of a concordance or index. Input for such noninfringing
purposes should be exempt from copyright.

(2) Input should be regarded as being merely the means of making a work available to users, i.e., as being comparable to the noninfringing act of placing a copy of a work on the shelves of a library.

(3) Even when a work is input for the purpose of making it
available for output, its output may never be requested.
(4) Input of itself does not affect the publisher's market
for copies of the work.

(5) The copyright license fees payable to the publisher
should be based on the volume of output. No separate
fee should be charged in addition for input.

In refutation of those arguments, and in support of the proposition that a license should be obtained before input, the following contentions have been made on behalf of authors and publishers:

(1) Input for any purpose entails the machine-readable reproduction of the work. Such reproduction and input of the work constitute a valuable use of the work, whatever the purpose may be. There is no valid basis for exempting such reproductions from the exclusive right of the copyright owner to make copies of his work.

(2) Libraries are generally expected to buy copies of the published works they place on their shelves. Likewise, computer systems should be expected to obtain the

machine-readable copies they need for input, or to obtain licenses to make them, from the publishers. If free input implies that computer systems are free to make their own machine-readable copies, the publisher's potential market for such copies would be destroyed.

(3) When output is contemplated, input of itself, by making copies of the work available as output, displaces potential sales of printed copies of the work.

(4) Licensing before input is necessary to enable the publisher to know that the work is being used in the system and to see that appropriate arrangements are made to compensate him for such use.

(5) Since a license will admittedly be required for output, practical considerations dictate that the terms of the license, including the basis for assessing fees, should be settled between the parties before input is effected.

As may be perceived from our earlier discussion relating to the input of data bases, in section A.4.4.3.4, we are inclined to believe that the weight of the argument comes down on the side of requiring licenses to be obtained before input.

A.4.6 BLANKET LICENSING AND COMPULSORY LICENSING FOR REPRODUCTION OF DOCUMENTS

A.4.6.1 Need for Blanket Licensing Mechanism. The ideal of providing researchers, through computerized data base systems, with bibliographic data relating comprehensively to all the published documents pertaining to any particular fields of science and technology has been mentioned in section A.4.2.2 of this report. Also mentioned there and in section A.4.2.3 is the further need of the researcher to be able to obtain expeditiously copies of the documents he identifies as being pertinent to his inquiry. And we noted that the documents needed for scientific and technical research are now mainly articles published in journals.

If and when computer storage of documents should become practicable on a sufficiently large scale to comprise complete libraries of virtually all the documents in any subject area, there will be a compelling need for some mechanism that will facilitate obtaining the licenses required for input and output of the mass of copyrighted documents in such a comprehensive library.

Meanwhile, the problem of supplying researchers with copies of documents on a comprehensive scale through other, existing sources, including libraries and other information centers, is already with us. (We have suggested earlier, in passing, that the time may not be too far off when such document supply centers will be operated in conjunction with, or as adjuncts to, computerized data base systems.) A few commercial organizations now supplying copies of copyrighted journal articles have succeeded in arranging for licenses from a large number of publishers. Libraries have been supplying photocopies of articles from journals in their collection but, with respect to copyrighted material, they have usually purported to do so within the limited scope of fair use.

It is generally recognized that, for a document supply center wishing to provide copies of articles from a large number of journals, the process of seeking out, and obtaining licenses individually from, each of the many publishers involved could be so time-consuming and costly as to be impracticable. (At any rate, this is the widely and firmly held consensus notwithstanding the success of at least two commercial suppliers of copies of journal articles -- University Microfilms and the Institute for Scientific Information -- in obtaining such licenses for a large number of journals.) It is also generally agreed that the publishers of copyrighted journals are entitled to be paid for reproduction of their articles (except for the limited reproduction permitted as fair use).

[ocr errors]

With two objectives in mind namely, to facilitate the mass licensing of copyrighted material for reproduction by document supply centers, and at the same time to provide for compensation to the publishers it has been urged that "clearinghouses" be organized through which blanket licenses could be obtained for an entire catalog of the copyrighted journals of as many publishers as can be brought within the organization, and lump-sum payments could be made for distribution among the publishers.

There are two existing types of blanket licensing mechanisms in other areas that might serve as prototypes for the blanket licensing of reproduction of copyrighted journal articles. One is a voluntary type of clearinghouse established by the copyright owners of musical compositions for licensing public performances. The other is a compulsory license plan established by the new copyright statute to permit the use of copyrighted works en masse, upon payment of lump-sum royalties, by CATV systems, jukebox operators, and educational broadcasters. We shall now look at these two types of blanket licensing mechanisms in turn.

A.4.6.2 Voluntary Clearinghouses. Possibilities for establishing a voluntary clearinghouse for the blanket licensing of copyrighted journal articles for reproduction have been under discussion, off and on, for a number of years. The development of an acceptable plan has been found to be beset with many difficulties. Two or three fairly detailed plans have been proposed in outline and put aside as unsatisfactory. The discussions so far have hardly gone beyond attempts to explore some of the possible bases on which such a clearinghouse might be organized and operated, and to expose the difficulties that might be encountered in establishing a workable mechanism.

A.4.6.2.1 ASCAP and BMI as Models. In the discussions referred to above, the clearinghouses operated by the American Society of Composers, Authors and Publishers (ASCAP) and Broadcast Music, Inc. (BMI), have frequently been cited as possible models that might be adaptable for the blanket licensing of reproduction rights in journal articles.

ASCAP is a voluntary membership association of writers and publishers of copyrighted music. It was established to license and enforce the rights of its members collectively in public performances of their music. A few statistics taken from recent reports will indicate the size and effect of its operation. Its membership consists of about 18,500 writers and 5,300 publishers of music. Its catalog of musical compositions is constantly growing, and the number of compositions covered by its licenses (a figure that is not announced) must now be well in excess of a million. Its gross revenues from domestic licenses is now over 80 million dollars per year, and from foreign licenses is over 13 million dollars per year. Its cost of operations in recent years has run to about 19 or 20 per cent of its gross revenues. The remainder of about 80 per cent is distributed among its writer and publisher members under a rather complex formula in which the principal basis for allocation is the estimated number of performances of each member's works.

ASCAP issues licenses to a number of different classes of users. The largest users, from which it derives a major portion of its revenues, are the radio and television networks. Other classes of users include local broadcasters, music and dance halls, orchestras and bands, hotels and restaurants, wired music services, business establishments, etc. ASCAP announces periodically a schedule into which its users are divided. As required by consent decrees of the United States District Court for the Southern District of New York, it must license all qualified applicants, all licensees in the same class are charged the same fees, and any licensee or applicant may request the Court to review the fees charged.

« iepriekšējāTurpināt »