Interesting People mailing list archives
IP: H.R. 2652 -- Implications for Science and Academic
From: Dave Farber <farber () cis upenn edu>
Date: Thu, 14 May 1998 05:14:59 -0400
To: USACM () ACM ORG
The attached analysis of the bill AS REPORTED and which would go to the House floor was done by Prof. Peter Jaszi of AU Law School. H.R. 2652 -- Implications for Science and Academic Research Although the "Collections of Information Antipiracy Act" is designed to address predatory competition in the commercial database market, the actual reach of this bill is far broader. As presently worded, it creates new rights in all kinds of data and has the potential to chill many forms of information use. Because of its overbreadth, enactment of this legislation would have serious adverse consequences for providers and users of data in science and the academic community generally. These consequences should be taken fully into account before any further legislative action is taken on H.R. 2652. Data is the life blood of academic research, and access to more data (be they records of physical observations or historical discoveries) is better than access to less. Under current copyright doctrine, decisions about what information to include in a collection or how to arrange that information may be protected -- no one can copy the manner in which a particular compilation is presented. Importantly, the data which make up contents of compilations are not subject to protection. Instead, they remain free for all to use. The current state of the law has been a positive one for scientific research. At it now stands, that law provides researchers with no incentive to withhold information in order to create conditions of scarcity. Rather, it encourages every researcher to make his or her factual findings available, by means of publication, as part of a common pool of knowledge from which all those working in the field are free to draw. Maintaining such an open information environment is critically important to the progress of research and development efforts in the public and private sectors. H.R. 2652 now threatens this basic paradigm of data exchange, by providing unprecedented new legal protection for information. Under its provisions, anyone who collects information (whether self-generated or generated by others) may be entitled to assert rights against others for the unauthorized "extraction" or "use" of the collection's contents. The bill's prohibitions would apply far beyond activities which are commercially competitive with those of the original compiler of an information collection. Under H.R. 2652, a researcher could be required to obtain a license in order to republish all or some of another's data,
and
to analyze or manipulate it to reach research conclusions -- if the first compiler had taken the necessary steps to establish or anticipate a market in permissions. It is important to note that the prohibitions contained in H.R. 2652 apply to researchers in nonprofit institutions when they
extract
or use information from the published research results of other scientists and academics. These features of H.R. 2652 will have distorting effects
on
the way in which information is made available within the research community. The economic pressures which would be unleased by H.R. 2652 are ones to which, sooner or later, the research community is likely to be forced to bend. Easy though it may be to suggest that even in this projected new legal environment, researchers could choose to continue to make their factual findings freely available, it is unlikely as a practical matter that many would be able to do so, at least for long. In an era of declining grant revenues and increasing pressures for cost recovery, every research project necessarily will be regarded as a potential profit center, as well as a scientific or scholarly undertaking. Thus, researchers may have no choice except to take advantage of H.R. 2652's invitation to engage in comprehensive commercial exploitation of their findings. Moreover, since every scholar and scientist both consumes and produces data, H.R. 2652 is likely to create a kind of economic "chain reaction" within the research community; for every set of research results which is made available pursuant to license, other researchers will experience new costs of data acquisition, leading them (in turn) to consider what additional revenues they may be able to derive from the licensing of their own findings. Ominously, it is precisely the data which have the greatest immediately scientific, medical or social importance, and on which others depend more heavily, which are most likely to be subjected to this treatment. There are a number of additional problems with H.R. 2652, if the bill is viewed from the perspective of the education communities: -- There are no limitations or exceptions which are relevant to the issue of re-use of previously published data in a scientific or other academic context. Obviously, the kinds of data of interest to scientific and academic researchers would be likely to satisfy the threshold requirement of having been "gathered, organized, or maintained" through a "substantial" investment of money or other resources. -- The 15-year limit on protection will be largely unavailing: In the world of research few data users can afford to wait. Moreover, enactment of the bill would extend retroactive protection to compilations created less than 15 years before the date on which it became effective. -- H.R. 2652 identifies the use of "individual items of information and other insubstantial parts" of a data collection, as well as "nonprofit education, scientific, or research uses," as "permitted." However, these qualifications on the new right are of trivial significance to the research community. Scientists cannot be expected to practice the sort of parsimony in information use which would be required to qualify under the first of these headings. As to the second, the permission for nonprofit uses applies only where the use "does not harm the actual or potential market for the work." In other words, this later provision is a reverse restatement of H.R. 2652's standard for liability: while uses which do deprive the compiler of revenue are permitted, those which "harm to actual or potential market" are prohibited -- even if they are undertaken for legitimate non-profit educational purposes. -- The relief H.R. 2652 offers academic researchers from potential criminal liability (as well the allowance for remission of civil damages where a researcher has made a good faith error in believing his or her use to be permissible) does little if anything of practical importance to address the adverse consequences of data protection for the research enterprise. Finally, it should be noted that while the bill's provision that data rights do not "restrict any person from independently gathering information or using information" not obtained from a protected collection, this qualification on protection has little meaning for the research community. Among scientists, in particular, new work strives to incorporate and move beyond previous results, rather than merely to reproduce them. Significantly, H.R. 2652 does not impose any limitations on the availability of protection for collections which are the "sole source" of particular kinds of information. This is a matter of special concern because the results of every scientific experiment, every physical survey, and every historical research program have the character of a "de facto sole source" database; subsequent scientists and scholars will have no practical choice except to draw upon the preexisting data if their new work is to advance the sum of knowledge. The adverse consequences of the proposed H.R. 2652 for science and education could be avoided if the bill were revised to incorporate a true "misappopriation" approach, focussing on the regulation of unfair competitive practices among commercial firms providing data to consumers. This approach to the problem of data protection would draw on such legal precedents as INS v. AP and NBA v. Motorola, and would emphasize the essential factors of classical misappropriation analysis in cases related to data: commercial "free-riding" by the unauthorized user, the time-sensitivity of the information involved, and harm to the competitive position of the compiler. Legislation with this emphasis would resolve the identified problem of predatory commercial practices without producing unintended adverse consequences for the nonprofit sector. =================cssp=============
Current thread:
- IP: H.R. 2652 -- Implications for Science and Academic Dave Farber (May 14)