Lapas attēli

the author, are not protected against use in the independent work of another author. In other words, it is only the author's original "expression" or exposition that is protected against copying. Copyright does not preclude others from using the know-how they learn from a copyrighted work in their own works. Thus, in the case of computer programs, copyright would not protect the processes or techniques developed to make the program operative and revealed in the program. This is reflected in the provision in section 102(b) of the new 1976 Act reading:

"In no case does copyright protection for an original work
of authorship extend to any idea, procedure, process, system,
method of operation, concept, principle, or discovery, regard-
less of the form in which it is described, explained, illus-
trated, or embodied in such work.

The protection afforded by copyright against reproduction may be of little or no significance with respect to programs designed specially for a particular user. Such protection may be quite valuable, however, for a program that would have a market of many users and could be reproduced cheaply in the absence of copyright.

[blocks in formation]

A. Copyrightability. Data bases are compilations of data consisting typically of bibliographic indexes words and phrases identifying the subject content of published documents -- and abstracts of documents describing their subject content more fully. Data bases may also consist of compilations of factual data such as mathematical or scientific formulas or statistical tables. Compilations of various kinds of data are traditional subjects of copyright protection. Both the Copyright Act of 1909 (in sections 5(a) and 7) and the new Act of 1976 (in section 103) mention compilations explicitly as a category of copyrightable works. In section 101 of the Act of 1976 a "compilation" is defined as "a work formed by the collection and assembling of preexisting materials or of data that are selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship."

As reflected in this definition, the authorship that makes a compilation copyrightable lies in the labor, skill, and judgment involved in

selecting the pertinent data and organizing and arranging the mass of selected data into a systematic and useful whole. Thus, while the individual items in a compilation are not subject to copyright in themselves, the collection as a whole, or any segment of it large enough to be the product of selection and organization by the author, would be protected by the copyright against unauthorized reproduction.

Compilations of various kinds of data including bibliographic indexes and abstracts -- are well known as printed publications and have generally been copyrighted in that form. A number of them are now being issued also in machine-readable copies and this trend seems to be growing. It is now possible also to compile indexes and other data by the use of computers, and there is no apparent reason why a data base so compiled, in machine-readable form, would not be copyrightable.

As reported in the February 1977 issue of Information Action (a publication of the Information Industry Association): "the number of data bases available for on-line access has doubled in the last year... In 1965, 24 machine-readable, bibliographic data bases covering 880,000 documents existed. In 1975, the total was over 160 covering 46 million documents.'

Many of the existing data bases are covered by copyright but others are not. Several of them have been produced by the U.S. Government and are therefore not copyrightable. Some producers of data bases apparently rely upon their contractual arrangements with the systems to which their data bases are leased for protection of their proprietary rights.

A. Copyright Notice on Data Bases. In order to maintain copyright protection, the published copies of a work are required by the statute to bear a notice of copyright in a prescribed form, "affixed to the copies in such manner and location as to give reasonable notice of the claim of copyright" (Act of 1976, section 401). Some commentators have anticipated difficulty in meeting this requirement in the case of machine-readable copies such as magnetic tapes. Their concern on this score may have been due in large part to the less flexible language of the notice provisions in the 1909 statute (section 20) which was phrased in terms of printed publications. In any event, we see no real difficulty in affixing the required notice to the magnetic tapes (or other machine-readable copies). The notice could be incorporated in the system software so that it would appear in any printout. And even assuming that an eye-readable notice should be affixed to the tape copies, it seems reasonable to expect the tape copies, or a container in which they are housed, to bear an eye-readable label showing the title which identifies the work on the tape; the copyright notice could readily be placed on that label. It might be added that any special problems regarding the placement of the notice on tape copies could be resolved under the Act of 1976 by the Register of Copyrights who is authorized (by Section 401(c)) to prescribe "specific methods of affixation and positions of the notice on various types of works that will satisfy this requirement."

A similar problem concerning the copyright notice occurs when some part of a data base is printed out from a computerized system in response to a user's inquiry. It is not clear whether the notice would be necessary on each reproduction of a relatively small number of items in a data base. It is arguable, we believe, that the reproduction of a small part of the collected data is not such a published copy of the work as would call for the notice; and this argument would be more cogent where the subscribers to the computer system's service were informed in advance that certain of its data bases were copyrighted. If it is thought to be necessary or advisable to have the notice appear on each printout of any part of a data base, this appears to be feasible. The data base would normally be identified by its title in the printout, and the computer could be programmed to include the copyright notice in every printout of the title.

A. Deposit of Copies for Registration. Registration of a copyright may be essential to its effective enforcement against infringers. Under the Act of 1909, registration is a prerequisite to maintaining a suit for infringement (section 13) and it facilitates proof of the validity of the copyright claim (section 209). The 1976 Act has provisions to the same effect (sections 411 and 410(d)), and provides in addition that awards of statutory damages and attorney's fees (special remedies that make enforcement of the copyright more effective) are to be granted only when registration has been made (section 412).

To make registration, the deposit of two copies of the work as published is required under both the 1909 Act (section 13) and the 1976 Act (section 408(b)). That requirement has been met readily for printed compilations of data, and printed copies would apparently suffice for deposit where the compilation has also been produced as a data base in machine-readable form. But if a data base were prepared only in machine-readable form, the deposit of copies could be troublesome, or at least burdensome, if, as the 1909 Act has been thought to require, the copies deposited had to be visually perceptible. The resolution of this problem has been made possible by the provisions in the Act of 1976 (section 408(c) reading

"The Register of Copyrights is authorized to specify by
regulation the administrative classes into which works are
to be placed for purposes of deposit and registration,
and the nature of the copies or phonorecords to be deposited
in the various classes specified. The regulations may
require or permit, for particular classes, the deposit of
identifying material instead of copies or phonorecords..."

A. Supplements to Update Data Bases. Bibliographic data bases must be brought up to date from time to time by adding to them new index entries and citations for more recently published articles. Some observers have seen difficulties in complying with the requirement for deposit of copies with respect to such supplemental additions. Printed publications with supplements issued serially, such as looseleaf information services, are well known. The usual procedure for them has been to publish each supplemental issue as a new work in itself with its own copyright notice, and to deposit copies of each supplemental issue for registration as a separate work. Alternatively, an entire new edition of the work as revised to include the supplemental additions could be published, and copies of the new edition could then be deposited. Either of these procedures would seem to be feasible for supplements compiled periodically for addition to a data base, though the latter procedure of publishing an entire new edition may be expensive.

It might be noted also that when supplemental items are merged into a computer-stored data base, coverage of the new material by copyright might require changing the year date in the copyright notice appearing with the data base in its earlier form. But even if the notice is left unchanged, copyright protection of the content of the data base in that earlier form would not be affected, and this may be adequate protection for all practical purposes as long as the newly added material could not be used without some of the earlier material. When the volume of new material added by updating over a long period of time becomes a major part of the entire data base, reissue of the data base in a new edition might be found appropriate.

A.4.4.2 Compiling Data Bases

A. Bibliographic Indexes. The process of compiling bibliographic indexes involves the following steps: obtaining copies of the documents to be included in the index, scanning those documents and selecting from them the key words and phrases to be listed in the index as subject headings, perhaps inserting other subject headings judged

by the compiler to be needed as cross-references, and arranging the subject headings together with citations to the documents in an alphabetical or other orderly arrangement. Traditionally, this process has been, and generally still is, performed manually through the exercise of human effort and skill, and the completed index is published in printed form.

It is now possible to perform this process and prepare an index of some quality by using a properly programmed computer, but with this difference: The documents to be indexed must be in machine-readable form to be processed by the computer.

As long as the indexer uses authorized copies of copyrighted documents, there is ordinarily no copyright problem in the manual compilation of a bibliographic index. Scanning of the copies, the extraction of key words and phrases as subject headings, and the arrangement of those headings with citations to the documents, do not constitute infringement of the copyright. No copy of the substance of the document is made in this process, nor would the resulting index be considered an infringing copy or derivative work since it would not convey the essence or meaning of the work embodied in the document.

Similarly, if a machine-readable copy of a copyrighted document used for indexing by a computer was obtained from the publisher,* preparation of the index by the computer would seem to involve no infringing act. A publisher who supplies a machine-readable copy of a work to a computer operator would impliedly authorize the use for which it was intended: Its input into the computer. The subsequent processing of the document by the computer in indexing it would be the same in character as the processing done in manual indexing, which, as pointed out in the preceding paragraph, would not involve any infringement of the copyright.

When a machine-readable copy is made available by the publisher, it would seem reasonable to expect the computer operator to acquire such a copy for his machine indexing. But if, instead, he chose to make his own machine-readable copy (which would seem to be unlikely since making

* The references made here and below to the publisher as the supplier

of copyrighted material assume that he is the copyright owner or the agent of the copyright owner.

« iepriekšējāTurpināt »