Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 282
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS 13 Promoting Access to and Use of Not-for-Profit-Sector Scientific and Technical Data—An Assessment of Legal and Policy Options MS. WILLIAMS: I am Martha Williams, and I will serve as the moderator for this session. I am from the University of Illinois, and I have been involved in databases my whole professional life and have sat on almost every side of this issue. I have been in the university. I worked for a not-for-profit organization at a research institute for 15 years. I have done a lot of consulting for industry, government, and not-for-profit organizations. I was chair of the Board of Engineering Information for 12 years. I was president of the National Library of Medicine for a time when Elsevier was suing the National Library of Medicine for their pricing policies and things of that sort. So, I have been in the database business for a long time. I also have a company of my own, the Creative Database. So, I enter the database discussion from the commercial provider, academic researcher and teacher, and the federal government point of view. In this session we are charged with addressing seven issues raised by the National Research Council (NRC) study committee (see Box 13.1 for a list of these issues). We want to address each issue from several points of view, the first of which is to be the legal viewpoint. We were to have two attorneys. Unfortunately, due to the weather, we only have one attorney, and I have asked him if he can present two sides of this. Our panelists are Jerome Reichman, who is professor at Vanderbilt University School of Law; from the provider point of view we have Richard Kayser, chief of Physical and Chemical Properties Division, National Institute for Standards and Technology; James Lohr is director of Information Industry Relations, Chemical Abstracts Service; and Leslie Singer, president of ISI. Our reactions from the user perspective will be from Stephen Berry, professor of chemistry at the University of Chicago. Box 13.1: Issues for the Discussion Session on Promoting Access to and Use of Not-for-Profit-Sector Scientific and Technical Data—An Assessment of Legal and Policy Options Should databases created by state universities be exempt from protection under new database intellectual property rights (IPR) legislation, or should they enjoy the same privileges as private universities? Why or why not? Who owns databases produced at a university? What are the determining factors? Assuming that government data will remain exempt from IPR protection, what should be the scope of that exemption for databases created with government funding at universities? Should all S&T databases created with government funds, either under contract or grant, be similarly exempt? If not, what should be excluded from that exemption and why? At what level of government funding should databases created at universities continue to be exempt? For example, should such exemption apply only for databases created with 100 percent funding? For databases created with at least 50 percent government funding? What are the relevant or determining factors? Who legally owns such databases?
OCR for page 283
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS Under what terms and conditions should the databases created at universities with government funding be made available? What are the salient factors that need to be considered? Aside from other laws/policies that may restrict dissemination (e.g., privacy, national security), should there be any restrictions on who can access the government-funded databases (i.e., category of user such as immediate research community, not-for-profit sector, commercial user, foreign user) and can discriminatory pricing be allowed? Why or why not? What principles and policies should govern interuniversity (or other not-for-profit) data access and use provisions? If stronger database IPR laws are adopted, should universities treat other not-for-profit users more favorably than the law requires? Why or why not? What kinds of rights should be transferred regarding data from databases owned by researchers/universities to publishers? In any new IPR regime what should be the scope of exemption regarding research, educational, library, and other “public-interest” uses of data sold by commercial publishers/vendors with respect to sharing and use, including transformative uses; integration into multiple-source data products or databases; dissemination to other parties within their own institution, or to other universities or not-for-profits; liability issues, generally; and other issues? What other important legal/policy issues should be considered in promoting access to and use of not-for-profit S&T data for the public interest? Examples of new legal provisions/policies for consideration include the following: Legislate the “first sale” doctrine in networked electronic environments in instances where technology allows no more than one user of a purchased intellectual work at a time. Legislate a depository library concept in which publishers to gain certain benefits must provide a digital copy of intellectual works and data sets to a national online collection that would then be accessible from public libraries across the nation. [See Geoffrey Nunberg (1998), “Will Libraries Survive?” The American Prospect, Nov/Dec, pp. 16-23.] Libraries might then “check out” these works to patrons as indicated in (a) above. Alternatively, if a licensing paradigm continues for access to online scientific and technical data, a portion of fees collected might be set aside (taxed) to subsidize access for schools and libraries in rural or underserved communities. (See Nunberg, 1998.) Development of standard licensing provisions and policies by libraries (see e.g., http://www.library.yale.edu/consortia/icolcpr.htm). Development of university policies that mandate that professors and researchers must maintain full nonexclusive copyright/sui generis right in any works or data sets developed in their capacity as university professors or researchers. Establishment of a “collection society” for S&T data. Other approaches? The first issue, which will be addressed by Jerome Reichman, is whether databases created by state universities should be exempt from protection under new database intellectual
OCR for page 284
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS property rights (IPR) legislation or should enjoy the same privileges as private universities, and if not, why not. MR. REICHMAN: That is really more a government data policy question than a not-for-profit question. I am uncomfortable with discriminating against the state universities; I think that all universities should be able to operate on the same basis. I think the federal data collected or generated by the federal system is what we are really concerned about in this panel. The problem at the state level where the taxpayer pays twice if they are state-generated data and if they are used by the state university with acquisition of proprietarial rights. Yet, we already do discriminate in copyright law. Federal government agencies cannot own copyrights, but state government agencies do own copyrights and exploit them all the time and survive. So, were it to happen that state universities were able to protect local data, it seems to me that that is a state issue rather than a federal issue. If the State of Tennessee does not want the University of Tennessee to own or protect the data, then that is something the state should say, but I don't know that they should be forbidden from protecting the uses of state data as a matter of federal policy under a new intellectual property regime. I guess in the interests of fairness I should say that my overall view is that state universities wear two hats. They wear the hat of not-for-profit user, but they should also be able to engage in commerce side by side with the commercial sector with a database that they want to exploit that way. State universities should be held to certain rules of fair dealing with other universities and the general rules that would be applied to all not-for-profit organizations. I think it would be unfortunate if a state university couldn't commercialize a database because the data were covered by state law and somehow they had been treated unfairly. But you can also take a contrary view and if there are limits on what commercial providers can do with federal data, maybe they should also apply to state data. I want to examine the general principle embodied in Office of Management and Budget (OMB) Circular A-130. Justin Hughes made the same point yesterday that taxpayers shouldn 't have to pay twice for the same data. However, there seems to me another context where that happens all the time. For example, where the National Science Foundation (NSF) funds a grant for research, the taxpayer pays for the research, but he also has to pay for the journal that publishes the article that describes the research results and the data. Is there something special about data that pulls against this result? I sometimes have trouble with the OMB A-130 concept, and I am not sure that is a universal principle that we ought to carry, or how far we should be carrying it. PARTICIPANT: That is one point I hear. Another point is that won't the state get back in taxes something from the exploitation that is made of those databases? DR. SAXON: Other people have raised that point, although that is a very appealing generic statement that probably sounds good in an advocacy congressional context. Perhaps our discussion will include a real examination of what it means about people not paying twice for data collected by the government, and what are the practical implications in any real situation for the reason you just mentioned. DR. BERRY: I think the question should be focused on not who supports the university but who pays for the research; and since research in state universities is supported by federal agencies, I think the criteria should be based on the fact that the support is coming from federal agencies, not from the fact that administrators and faculty are paid by state funds.
OCR for page 285
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS DR. LOHR: I almost from the outset need to disqualify myself from this discussion. This is kind of an esoteric subject, but it seems to me for some of the reasons that have already been given for administrative simplicity, for basic fairness, however people are treated in this context it ought to be the same. DR. KAYSER: The only thing that I would add is if a private university wanted to develop a sweat-of-the-brow type of database, it depends entirely on the source of support. If they wanted to develop such a database with their own source of support, that is fine. If a state university wanted to develop a database with their own source of support, that would be fine as well. The only difference would be that the state might have a right to say something about the use of that database depending on what kind of data were included. I don't think the federal government should tell the states what to do. MS. WILLIAMS: We will go on to the next question then: Who owns databases produced at a university and what are the determining factors? MR. REICHMAN: That is a complicated topic that really isn't terribly germane to the subject at hand. This has to do with what the university's copyright policy and patent policy is, and then it has to do with the vagaries of the federal patent and copyright policy, which does not always or clearly tell you who owns what. For example, even in patent law, which is a federal subject, ownership is regulated by state law. To find out who is the owner of a patent, the inventorship is regulated by state and by federal law, but the ownership is governed by state law. The universities have these historical rules that if something was invented within the scope of employment the university would normally take the patent and give the inventor a share of the royalties. On the other hand, if a professor's work was copyrighted then it was thought that the professor should own the copyright, and the university shouldn't have that. Of course, modern intellectual property rights have skewed all that because they deviate from the historical patent-copyright dichotomy. For example, for reasons of pure expedience and little logic, computer programs have been put under copyright law. So now, a university technician who creates a computer program could be more favorably treated than if he or she had patented the program. As a result, many universities have revised their policy so as to claim the copyrights in computer programs and treat their creators on a par with their other engineers who develop patented inventions. Basically, the gap or problem as I have defined it is know-how. Know-how isn't generally governed by any federal regulation (except for criminal laws on industrial espionage). The legal regime governing know-how is actually tort law concerning misappropriation in general, and misappropriation of trade secrets in particular, which is state law with its own rules on ownership. The university has to step into this void and negotiate its own rules on who will own what and what will happen in a sector in which the legal ownership rules are very volatile. This would be the situation of databases: Ownership would be unclear in the absence of a university rule or contractual provision. As a general principle, an investigator who generated data with university resources and within the scope of employment, should be contributing something to the university if commercial exploitation occurs. But I am not sure why these ownership issues regarding university databases would fit in this breakout session. PARTICIPANT: This question does touch on whatever exemptions we decide shouldn't apply for university-created databases.
OCR for page 286
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS MR. REICHMAN: What you would have to say is that a university will have to review its policies immediately with regard to the ownership of databases. DR. SAXON: But if I understand you correctly, what you are talking about as the university, which is a not-for-profit organization, is how the university is going to view databases. That is independent from any legal regime how the university views databases. MR. REICHMAN: That is right. The legal regime won't tell the university how to resolve the ownership issues. DR. SAXON: The point we can distill out of these remarks is that any proposed legal regime doesn't really address this particular question of who owns databases produced at a university. DR. LOHR: I wanted to add that if any of us go to work for a private employer to create intellectual property and we are successful, there is a property right that comes along with this. Almost invariably we will have signed a paper the day we walked in the door that says that property right belongs to our employer. On the know-how side if we have learned things along the way and we decide to quit or get fired, occasionally key employees are enjoined from working for competitors for a limited period of time and using that know-how, but it is frankly kind of rare. I don't think that is a bad principle to apply here. MR. REICHMAN: Universities generally have not done that. That kind of restraint on using your knowledge, would raise a lot of problems. DR. LOHR: I am not suggesting that, but I am thinking more in terms of the property rights if there are in fact any property rights that grew out of any work, the institution will own them, and it can do what it wants with them. DR. BERRY: I think we have to make a very big distinction here between the not-for- profits and the for-profit institutions. The ground rules have to be completely different and here ownership is perhaps not as precise a term as I would like to use as the rights to distribute. For the not-for-profits and at least as much for the users, mostly users even more than providers, the issue is one of access. I would like to come back to the issue that most university research is supported by federal agencies because the government believes it generates a public good. To achieve the benefits of that public good, the information must be distributed. In fact, this is a special kind of public good because if you contrast it to the classic economic good, whose value does not diminish with use, the value of scientific information increases with use. So, it is at least implicit in the intent of the government supporting the research to have maximum distribution of this information in the sector that generates public good. It is in no way incompatible with the government allowing the private sector to use the public data to generate a private good, as long as that doesn't inhibit the public-good aspect. The intent of the supporting government is best served by the policy that encourages maximum distribution of the information and discourages or prevents any inhibitions to that distribution of the data and databases that are generated by federal funds. Consequently I think that has to place a constraint, a very severe constraint, on any commercial distributor of government-supported information. DR. LOHR: I understand your point, and I think it is a good point. However, how do you accomplish the end that you decide is the best? Here we are talking about who may own the thing and what the owner may do. It seems to me it is better to have what you just said codified into a very understandable policy of the granting agency. If I am a university, and I have an internal policy that I am determined to take everything that comes out of that university and
OCR for page 287
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS commercialize it for top dollar, they just start withholding grants from me and deal with it that way. DR. BERRY: In fact there was a group of researchers who published a Policy Forum in Science magazine proposing something that the American Chemical Society (ACS) hates, that the authors of scientific publications supported by federal funds keep the copyrights to their articles and only give licenses to the journals. DR. STEFIK: I have a question on that attitude about the restraint on the people distributing data; in particular, whether the economics has to be entered into the equation of the discussion. As we heard in the commercial data panel yesterday, most of the commercial publishers of databases focus on where they add value to the data to make the data more usable. Now, one doesn't want to hinder the commercial exploitation of otherwise perfectly usable raw data. DR. BERRY: I think you are quite right that when commercial publishers can add value sufficient to justify the user paying for the commercial database, that is fine as long as the original public data remain available. You also have to recognize that the commercial producer has a responsibility with each new potential venture of assessing whether that is going to be profitable, and it is the responsibility of the commercial publisher to decide whether that is worth publishing or not. There is no reason in the world why every database should be a natural venture for a commercial publisher. Many of them will not be profitable. In those cases, in order to achieve the public-good goals, it is the responsibility of the government through its research agency to see that the original information is distributed. In other words, it is the responsibility of NSF and the National Institutes of Health (NIH) to be sure that, for example, the page-charge costs are included in the grant. It is a tiny amount added to the grant, but the funding agencies should not assume that every piece of research information is going to find a commercial source for publication. I read about a new journal that will be free on the Internet. The cost will be paid by the authors. I don't know whether it is going to be a successful experiment. There will be lots of experiments of that kind, but I think that we are going to see many new models experimented with in the coming years, and that kind of thing is one of them. DR. STEFIK: I was taking up a different aspect of what you said, maybe not understanding your intent. I saw what you were saying as being a restriction on what people could charge for value-added databases. DR. BERRY: I am all for value-added databases. DR. STEFIK: Let me add another wrinkle on that same thing. As we heard yesterday, a great deal of the value in the use of databases is when they get combined. This strikes me as being analogous to the operation where we think about reading journal articles and combining ideas. For databases, the combination takes place in the computer. If there are lots of licensing restrictions on any particular database, it becomes harder and harder to trace the audit trail to the different data that might be combined and possibly get some results. I am wondering whether we are prepared to handle a great deal of complexity in licensing ability and similar restrictions with this combinational activity taking place. This certainly makes a case for real simplicity in terms of being able to combine data from different sources. I am not at all in principle against some wonderful new economy, but I cannot imagine what it would be. I haven't seen any mechanism for this, but I think there is a reality here. If the real value comes from combining databases, we want that to happen and we ought to pay attention to that goal and not put too many restrictions on it.
OCR for page 288
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS DR. KAYSER: I would tend to take a really narrow view of this. It seems like most of what we have been trying to focus on in this workshop is what you do with data collections produced by sweat of the brow. Then the key point is what intellectual property model do you want; and if you go to the sui generis model it seems like that opens up a huge can of worms as to who owns the data and who can use them. On the other hand, if you go to a misappropriation model, then I think that does the least harm, or that would require the smallest number of changes compared to the status quo. I would say that what happens to this information in this model really becomes a question of how was it supported, and if it was supported by the government under contract it goes into the public domain. I, personally, think that the government should not allow grants to be used to support the development of proprietary sweat-of-the-brow collections either. So, I would say that all such information generated using government funding should go into the public domain. DR. BERRY: I would like to make a little distinction between what you are referring to and what I had in mind; that is, when you say “sweat of the brow,” you are implying that the data already exist and are there somewhere in the literature, and all that is required is the compilation step. In my experience, I have seen very few instances in the scientific data realm that are really comparable with Feist. It has always seemed to me there is a much more intimate relationship between the data generating process and the data compilation and distribution process, so that the sweat-of-the-brow argument that was made in Feist is, I think, simplistic to describe what we usually use in our own research. In fact, one question that I would raise here is, Would it be correct to say that a very large fraction of the scientific databases that people use have some intellectual added values in the creation and in the presentation, enough so that most of them really wouldn't fit Feist but would be copyrightable? MR. REICHMAN: The Register of Copyrights correctly cautioned not to get hung up on the eligibility standard of Feist, although that does become relevant with these machine-made data. The real problem with Feist is that there is no scope of protection for the data as such. DR. BERRY: Take, for example, the Journal of Physical and Chemical Reference Data, which is copyrighted. DR. KAYSER: Whether or not the journal's copyright will stand up, though, is another question. That is part of the status quo, though, not necessarily what we are talking about changing with this database policy. If the journal is actually copyrighted, then everybody who is collecting information from the literature is violating copyright. MR. REICHMAN: They are probably taking the noncopyrightable components (including the data as such) and then there is an additional question of fair use if they took original elements of selection and arrangement (such as a set of tables), which do not attract copyright protection. DR. KAYSER: That is a question of where the line is between them. MR. REICHMAN: I want to focus a little bit on this university status that is addressed in the question. I just want to throw in some background from the Senate Judiciary Committee negotiations so that everyone is aware of the concrete issues with regard to government-funded data that were raised. We started out by asking, do you allow that anybody, disregarding the universities, can make a database in which there are government data? What can they do? Generally the negotiators were agreed that yes, the private sector could and should be encouraged to make use of government-generated data, especially value-adding uses of such data.
OCR for page 289
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS If that premise is true then the question became, What about rights for scientists and educators to the raw data that become incorporated into the now-privatized database? We feared that there was a serious risk of capture, and that there ought to be serious no-capture provisions. What sort of no-capture provisions should there be? We thought that a rule—and we actually had it written into the final draft, and I think we had substantial agreement on it—should require the private purveyor to keep and make available, if not otherwise reasonably available, the original government-funded data set to scientists and educators at a reasonable cost. It is a little different from what federal policy says, which is to make the data available to anybody. We said, at least make it available to not-for-profit scientists and educators, if the original data where unavailable elsewhere. If you think that the private sector can privatize government data by adding value, then why not universities? Should they be any different? If universities are going to have a privatizing function, then you could argue that there shouldn't be any difference except that they should be subject to the same no-capture provision. Then you could take it one step further and say, “but we should have a Hippocratic oath for universities.” That came out of our discussion that at least universities should develop their own, higher-level guidelines to guarantee what would be done with all data. Then, finally coming to your point, what ultimately happens to government data as well as to any data is affected by the type of intellectual property regime that one selects. So, if one takes the view that we are going to have either a strong exclusive property right or, as the Register of Copyrights conceded, a self-proclaimed misappropriation approach that is no different from a strong exclusive property right, then it becomes really important who owns what and who does what. If you have a very soft misappropriation regime that only kicks in if someone is going to take a wholesale chunk of the database, then your main concern is not to reduce the incentive to invest. There is a proposal circulating about a soft misappropriation approach, based on wholesale duplication, but I don't know whether it has a chance or not. Then you have to worry a lot less, but you still have to worry because the question becomes, can the university then sue for misappropriation (i.e., wholesale duplication) of its value-added database? I think the basic question with regard to universities is, what you are going to do about the no-capture provision. MS. WILLIAMS: Explain “no-capture.” MR. REICHMAN: The no-capture that we are talking about is that you cannot capture the federally funded data so as to become a sole-source provider of those data subject to no limitations on excluding others. MR. ASHLEY: I have a question about the no-capture provision as applied to especially federally funded scientific data, and I am just trying to put my finger on what the risk is. What is it that would change under a new regime that is not already happening? We have private distributors of federally funded data now who presumably can essentially control the government data in a way that we might find undesirable, but I am not really sure why that would change if you changed the regime. Is it a question of providing more money for government archiving functions of scientific data, or is it a question that we get recurring through the debate, as you know, of why they can't just go back to the underlying data? MR. REICHMAN: All those are good questions. The only thing that changes is the potential potency of owning the data and the potential monopoly that comes from the exclusive property right or the high protectionist misappropriation approach. So, now, in addition to the
OCR for page 290
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS fact that you had the risk of de facto capture before, you now have a legally enforceable monopoly, and the risk is that nobody else can ever get the data to overcome a formidable entry. Now, what could the government do? The government could avoid that monopoly by certain contractual provisions or by archiving measures when technically feasible. When we were in negotiations, we convinced the legislative staff that the proponents of the legislation were trying to get perpetual protection in the revised bill as can occur in the European Union. We proposed a clause clearly saying that protection had to end; if you are going to use this high protectionist model, it has to end after 15 years, and there should not be any possibility of perpetually renewing the term. Some participants also suggested that a deposit system was needed because we won't be able to tell what is the first release of the database, which expires after 15 years. A dynamic online database will keep evolving and the producers will keep adding to it. The database producer will keep saying that he has further rights in the new additions to the database. We also wanted the database producers to deposit Release 1, so that at least after 15 years we could say, go on protecting your new stuff but Release 1 is in the public domain. That would be similar to the no-capture clause for government and operate as a kind of deposit, but instead of requiring a national deposit that would require additional federal funding, we argued to make each private licensee responsible for that archival function. Who will pay for that archival function? Can the publishers not recapture that? We said, yes. If you read the final Senate discussion draft, the database producers could charge reasonable delivery and maintenance fees for that service, but they had to make it available. As it was worded, they could not avoid having to keep a data set for science and education. DR. BERRY: There was a National Research Council study about archiving electronic scientific and technical data specifically requested by the National Archives and Records Administration because they wondered whether they should do it, and the recommendation of the report was no.1 The National Archives were absolutely inadequate for archiving electronic scientific data. They might be fine for old books and documents, but they were completely unprepared for the volume of scientific and technical data. So basically there is no central place to put the scientific data in government right now, except for the individual agencies, such as the National Library of Medicine. The U.S. Geological Survey (USGS) handles its data. The National Aeronautics and Space Administration handles its own. NOAA handles its own. But what they wanted us to say was that after a certain length of time they should give this to somebody else to take care of. DR. LOHR: It may well be that they are unprepared to take over that burden, and I understand your position, Professor Reichman. You have got to get the best deal you can when you are negotiating. As a practical matter, to create an environment that says that any private entity that wishes to use this government data is in the first instance obligated to maintain this file I think is a very, very risky policy, if, in fact, you really wanted to assure that the material exists over a long period of time, because companies come and go out of business. How would you deal with that simple case? MR. REICHMAN: You are right. It was second best. We wanted a mandatory deposit if there was a high protectionist regime. Now, that would meet your objections if you had a mandatory deposit, and then, assuming you could solve the technical and financial problems, then you would have a de facto national data archive. By way of contrast, if you have a simple 1 See National Research Council (1995). Preserving Scientific Data on Our Physical Universe: A New Strategy for Archiving Our Nation's Scientific Resources, National Academy Press, Washington, D.C.
OCR for page 291
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS misappropriation approach that only kicks in against conduct that genuinely affects the incentives and risks of creating databases, you don't need a 45-page bill and you don't need a $10 billion national data archive. DR. LOHR: You might well very shortly have a $10 billion national archive. First of all, those data accumulate at a furious rate; and second, the technology of data management and storage changes so rapidly, probably every decade, if not sooner. So, you are going to have to take everything you have and convert it over to a different medium, and this is going to be big money real fast. DR. SAXON: I would like to take a look at the questions raised by the NRC study committee and try to pick out the things that we feel here we could make the best contributions to. DR. KAYSER: It seems to me that the whole issue is that there are certain collections of information right now that you could argue are not protected, and all we are talking about is changing the law in such a way that they are protected. You could do a lot of work putting something together in a database, and I could come along and rip it off, produce a product that looks very much like yours in direct competition with you, and that doesn't seem right. So, the question is how do you fix it, and there are two possibilities. One is that you create new rights that don't exist right now. I really like what Marybeth Peters said yesterday. However, in practice the two regimes may work out the same way, but you can either create new property rights or you can go the misappropriation route. Going the misappropriation route seems to be the one that would help us avoid all these cans of worms because I don't think anything is broken except in this potential for rip-off. MR. REICHMAN: There is large support among the scientific community for a misappropriation approach, but you have to be careful. Just saying the word “misappropriation” doesn't cut it; Ms. Peters talked about the so-called misappropriation approach that Congressman Coble introduced. That is a high-powered property right by the back door that has been unmasked. So, just saying, “misappropriation” isn't enough; what you mean is a minimalist misappropriation approach as opposed to a maximalist misappropriation approach. DR. KAYSER: You have to decide from the point of view of policy options. First, you have to decide whether you want to create a new right or whether you want to go the misappropriation route. If we decide the latter, then we could start worrying about the details, but there is no point in doing that until we decide what route we want to take. MR. REICHMAN: During the Senate negotiations, we discussed that calmly, in abstract. I think the dominant group of proponents is taking the demonstrable minimal risk of market failure as a pretext, as an excuse to get a maximum right. DR. BERRY: I think there is an important point, another way of saying what Jerry submitted, which is that the same word is used really for two completely different policies, and the policy that Jerry is trying to avoid I can put in very different terms. The maximum misappropriation policy would be one in which the legislation provides a higher level of protection than copyright for something that, according to Constitution and Feist, at most deserves a lower level of protection than copyright, and this, I think to many of us, is the bitter incongruity of the past two attempts at enacting this legislation. DR. KAYSER: I see the sui generis model as a bad road to go down. Once you are talking about misappropriation, you can talk about strong misappropriation and weak misappropriation. I think what we want is weak misappropriation. DR. BERRY: That is right.
OCR for page 292
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS DR. KAYSER: In our discussion we should try to draw some conclusions that support at least one of the options. MR. REICHMAN: Questions 4 and 5 posed by the committee have some relevance (see Box 13.1). They are saying, Should universities treat each other better than they would treat others? In other words, in the law there would be some carve-outs for science and education. Naturally we have been arguing for very strong carve-outs from a very strong property right. That is a real issue, but whatever the carve-outs are for science and education, then the question would become, Should universities have a set of guidelines among themselves governing interuniversity use of data generated in universities that may give universities higher privileges (and fewer exclusionary powers) than are available under the law? Second, should universities have a concrete set of rules that people dealing with universities would have to accept—a common licensing policy, etc.? I, personally, think that is a good idea. During the negotiations, someone cautioned us to be careful because university agreements on transfer of technology licensing broke down. That is a true statement, and a very fair comment. By the time you are talking about transfer of technology agreements, you have the technology. Anybody can see the value of it. That is hard to discipline in any world. We intellectual property lawyers know that. However, with respect to databases, we are all worried about lost opportunities because we are going to put these rights way upstream. I think all universities have such an interest in keeping that upstream flow of data circulating and that they would self-enforce a deal that they made among themselves. DR. BERRY: I would like to believe you, but I am afraid there is some evidence that the expressed concerns regarding university agreements on transfer of technology licensing are real. We see what I think is the beginning of a terrible erosion of the kind of open atmosphere that universities have enjoyed. For example, two research groups in the same department have contracts with two different commercial firms to work on pharmaceuticals, and the students in one research group cannot talk to the students in the other, and this is actually happening. Another example that I found very scary, though reassuring in its outcome, was about a young woman who was interviewing for a job. She was asked a question about details of a process, and said, “ I am sorry, I am not allowed to tell you that.” Of course, what happened was she was no longer a candidate for that job. I think that this erosion of openness is a real problem. People are trying to start their capture very far upstream. MR. REICHMAN: It is true, and this intrusion of trade secret law into the academy is terrible. I just have a feeling that for data, this erosion is likely to be less aggressive because the data are so far at the beginning of the applications process that you cannot see the immediate return. However, the university can see the consequences of not being able to get at the data for its other departments. Don 't forget that there are lots of other enforcement mechanisms, such as granting agencies. In addition, when people start opting out, these holdouts are going to have to deal with the rest of the university community. If you have a university licensing authority to license all of the university data for all of the university, people who try to hold out would get hauled before the intellectual property committee, and I don't think they would hold out for long unless there were good reasons to do so. MS. WILLIAMS: I put a few notes on the board if I can just take two minutes of your time and try to get a little bit of structure to some of these thoughts. One of the things we are talking about is the source of dollars that go into data and databases. They come from the government, not-for-profits, and industry. Government data may go to not-for-profits and from them to industry, or the data may go directly to industry.
OCR for page 296
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS DR. BERRY: It has to be made maximally available. I don't think anyone is saying that it has to be made available free, but because it is produced as a public good it in itself should not available to be a cash cow. DR. RUMBLE: How do you differentiate that database from the human genome or some USGS data, which everyone says have to be freely available? DR. BERRY: Don't say freely available; openly available is different. DR. RUMBLE: You could say, available only for the cost of reproduction and dissemination under OMB Circular A-130. What is the differentiating factor? DR. BERRY: Add in the cost of adding the added value. It is a straightforward economic issue. DR. SERAFIN: It seems to me that these reports are not available freely today. If you want a publication from any respected journal, and you want to read the results of that research, you have to pay for the journal subscription or you need to go to a library, which has paid for the journal subscription. The authors have signed away their copyrights to the publisher. These are just facts, and it is my judgment that this has not stood in the way of effective communication of research results. I don't believe that we are challenging that aspect of our scientific and technical community in this workshop. Now, if in your judgment certain protective rights in the future would challenge that and change the way people had access to research results in a significant way, then I think we ought to try to identify that and argue how that would occur. DR. BERRY: I think it was made fairly clear yesterday that the paradigm you describe worked beautifully and there was a true symbiosis between the publishers and the scientists until the Internet came along. This whole new mechanism has changed the picture, and we are groping for a new set of rules in that new environment. So, I don't think that we can start with the premise that you stated. We are already confronted with having to make a change. MS. SINGER: I don't think that paradigm is changing all that fast. There are exceptions, such as high-energy physics, where perhaps the cohesiveness of that discipline allows the author to actually publish in the Internet environment and feel comfortable in his or her career and everything else. However, the great majority of information still goes through the traditional publication route, whether it is funded by the government or not. It just seems that that paradigm is still in effect where authors feel comfortable in a lot of disciplines for their career review, so that they sign the copyrights away to commercial publishers or not-for-profit publishers. DR. BERRY: Fortunately, the process has occurred slowly enough for us to think about it along the way. DR. SAXON: This discussion does bring up something that hasn't been a part of this workshop, but potentially might have been; we have been talking about more traditional forms of data like collections of numbers or gene sequences. The way the antipiracy act is written, many commentators have interpreted it to include collections of full-text publications. This group might have some commentary on how the broad interpretation of what we used to think of as print publications or primary publications will play out to the traditional norms of scientific exchange in a new regime. It is a different kind of exchange than making secondary and tertiary databases, but nonetheless is very crucial to the scientific effort. DR. KAYSER: I think the reason that the current way of doing things hasn't created a problem is because the data are freely available for anyone to use. DR. SERAFIN: I challenge that. Publications oftentimes do not have the data. Publications have the conclusions drawn from the analysis of the data. The raw data sets are generally sitting in the laboratories or the offices of the researchers, and they are not on the
OCR for page 297
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS Internet broadly. Sometimes little bits and pieces are available online and if you want access to those data you go to the researcher and ask the researcher for the data set. In some cases there are centers like the National Center for Atmospheric Research where certain data sets are collected and archived. People do research with the data sets and then we make them available to the community at the cost of reproduction. DR. BERRY: But I think the genome database led to a new paradigm in the genomics community, where articles are not publishable unless the scientists have submitted their sequences to the database. I think that was an extremely interesting change in the way we operate. DR. KAYSER: My main point was not necessarily that the data are available, but that they are not protected. If we went into a new paradigm where the data were protected and owned, for example, by the author or someone else, then that could change the paradigm. DR. STEFIK: The question is, Should government data remain exempt from protection? And there are a bunch of provisions here about the degree of funding. By the time you get to Part d of Question 3, should there be any restriction on who can access the databases and so on, and I think for the most part we haven't addressed these, and I think, also, it is appropriate. By and large, there are all these conditions when you get a government grant, and many of the ambiguities about what counts as data are pretty complex. We are not talking about lab notebooks. Those aren't going to be made generally available in any case. The point I am making is, we haven't had a principal discussion about those issues, and it is not clear that we have a lot to bring to that. MR. ASHLEY: For the record, I think there is sort of an implication of a one-size-fits-all solution to treating all types of government data, whether they are funded through grant or contract or what have you, and there are lots of complexities. There is a distinction between at least how the government treats data that are generated by contractors versus how they treat data generated by grants. I think the time doesn't permit us to go into those complexities. I think I agree that it is probably best to move on. MS. WILLIAMS: Let us go on to Question 4: What principles and policies should govern interuniversity (or other not-for-profit) data access and use provisions? If stronger database IPR laws are adopted, should universities treat other not-for-profit users more favorably than the law requires? DR. BERRY: I think the consortium model previously described by Professor Reichman is a very good answer. MR. REICHMAN: Universities need to discuss the consortium model. I don't think it is something you can impose on universities. You can only impose on universities to do what the law says. Now, the law should ensure that not-for-profit scientific and academic users are left in no worse a position after this database law is enacted than they are now. DR. SAXON: Our main focus is what legal regimes can serve the enterprise best, rather than dictating behavior among private parties. MS. WILLIAMS: Let us go on to Question 5. What kinds of rights should be transferred regarding data from databases owned by researchers and/or universities to publishers? I believe that is really a matter of negotiation. You give up what you are willing to give up. DR. BERRY: I would like to see the following tried as an experiment for the generators of the databases that have enough intellectual content to justify the consideration for copyright and therefore the generators by law de facto own the copyrights initially. The experiment is that the creators or generators of these databases who do have the copyrights keep the copyrights and
OCR for page 298
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS give licenses to publishers. This has been suggested for scientific articles, particularly those supported by federal research, in several different venues. The issue is whether licenses can be written that allow the creators of these databases to have the freedom to distribute the information as they would like and still allow the publishers to do everything that they do now except constrain the creators from distributing the information. MS. WILLIAMS: The authors are competitive though. DR. BERRY: That is part of the ground rules of the experiment. The creator of the database is not interested in being a grand-scale circulator. MR. REICHMAN: I don't think universities would agree with that scenario. The situation is more like software. Universities will probably feel that they put a lot of their own money, time, and resources into the creation of these database. DR. BERRY: They can share in the owning of the copyright. MR. REICHMAN: I think that the actual ownership rules may be more complex than that. DR. SAXON: It gets worse than that because if you take a literal question, it is rights regarding data. So, this goes back to the collection of facts issue and whether facts themselves are subject to intellectual property rights that do not follow standard interpretations of copyright. In other words, you are making the copyright analogy when you have a sui generis regime where you have facts now that under any other regime don't have protection. If I contribute the results of my research, which are these data, into a larger database, what rights do I retain? DR. BERRY: What restrictions should you impose on your publisher? DR. SAXON: Yes, and that goes back to the same question where you have metadata, how much do the restrictions of individual bits of it carry up/down through the chain. MR. REICHMAN: This comes to the issue of fair use. DR. SAXON: It is what rights reside in the numbers. MS. WILLIAMS: That depends on whether you are talking about raw data or reduced data. DR. BERRY: Under copyright you can; under sui generis, no. DR. SAXON: That is the issue, and I am not sure that we have an answer. DR. RUMBLE: The answer is, if you don't want to create new rights of protection, then you should not amend the law. MR. REICHMAN: And the other larger answer is that if universities are going to try to establish a common set of guidelines so they don't kill each other or academic research, this may be one of the things they discuss. However, this is not a thing that the federal legislation will resolve. DR. BERRY: Except insofar as any instrument specifically for databases should be at least as generous on fair use as copyright. MS. WILLIAMS: Let us go on to Question 6: In any new IPR regime what should be the scope of exemption regarding research, education, library, and other “public interest” uses of data sold by commercial publishers and vendors with respect to sharing and use, including transformative uses; integration into multiple-source data products or databases; dissemination to other parties within their own institution, or to other universities and not-for-profits; liability uses, generally; and other issues? MR. REICHMAN: This is the crux of it. We have to understand science is coming to these congressional negotiations in a defensive posture. The commercial publishers want a maximalist regime. In the course of the Senate negotiations, the commercial publishers clearly
OCR for page 299
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS said that if you have access to a commercial database and then take a chunk of it and put it into a new database that would address a new problem, it would be violating their redistribution rights. There are two issues. The first is something that we will call fair use, but it is not fair use. I prefer to call it a carve-out for science and education. The other is misuse, or restrictions on licensing for science and education. If you gain a carve-out, an exemption, but you don't put restrictions on what the publishers can do with contracts, they will simply contract away that exemption, which will be upheld by the proposed Uniform Computer Information Transactions Act (UCITA, formerly known as Article 2B of the Uniform Commercial Code ). Now, what is wrong with fair use? Fair use works fine in copyright law because there are built-in exemptions that are enormous, such as the idea-expression distinction, the defense of independent creation, plus a whole list of things that academics can do officially. In this database milieu, lost of what we used to do will become infringing use because everything that used to be free will now be owned. Therefore you don't want fair use. Fair use would mean you are constantly infringing with almost every scientific activity, unless you go to court for an exemption. A carve-out principle, based on the notion that science and education should be no worse off than they were before, assumes that customary traditional scientific activities are presumptively legal. DR. BERRY: I think Question 6 essentially clarified it for me, distinguishing fair use, which was technically a term to allow exemptions from what would be violations, to use a different term that I naively would have said, “Everything should be fair use, except these exemptions. . . .” The term “fair use” would be badly used there. I understand now what you are saying, but I think that what some of the publishers realize, and that is still very, very much a potentiality, is that the scientific community has the possibility for its own purposes of creating its own mechanisms for distributing information that wouldn't be in any way relevant to the present commercial publishing world. The alternative is for the scientific community to break out completely, follow something like the Los Alamos e-print archives model, and find its own internal ways of distributing the information. It sounds revolutionary, but it is not really because we know that in certain scientific arenas it already exists. DR. RUMBLE: But there is another model that is intermediate to that, that the institutions that hire the scientists to do the work create their own small empire. Stanford University has already created HighWire Press, which does electronic publishing and it just appears from the press releases and business announcements that they are in it to make a lot of money. I don't know for a fact, but I have been told that the University of California has seriously considered creating a University of California research journal covering all their institutions with the idea of making money that goes into the coffers of the University of California, and they could conceivably be interested in the same kind of strong protection that anybody else is who is interested in making money. DR. BERRY: But at the moment, the scientific community sees itself aligned with the university administration. MR. REICHMAN: For the record, that is not necessarily the case. The universities have played a most ambiguous role. If we could have brought the universities to the streets with 400 university presidents telephoning the White House last year, we would not have risked coming so close as we did to such a disastrous end. During the last session of Congress, the Association of American Universities, too, held a very neutral position. Universities are at risk here, but they also want to commercialize their products. They want a balanced approach that doesn't undo
OCR for page 300
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS them. So, officially the American Association of Universities has been on the public-interest side of the negotiations, but like a lot of other scientific organizations, especially many of the professional societies, they have not spoken up at the national level because they are conflicted internally. (Recently, however, they have become more active in seeking a balanced solution.) PARTICIPANT: The scientific membership societies that derive significant revenue from publications, like universities, find themselves on both sides. They should be the best spokesmen for a balanced approach. DR. SAXON: It is a complicated issue. MS. WILLIAMS: The universities want to have their cake and eat it, too; and it puts a lot of pressure in strange places within the universities. Let's move on to Question 7: What other important legal and policy issues should be considered in promoting access to and use of not-for-profit scientific and technical data for public interests? Examples of new legal provisions or policies for consideration include the following: Legislate the “first-sale” doctrine in networked environments in instances where technology allows no more than one user of a purchased intellectual work at a time. Legislate a depository library concept in which publishers who gain certain benefits must provide a digital copy of intellectual works and data sets to a national online collection that would then be accessible from public libraries across the nation. [See Geoffrey Nunberg (1998), “Will Libraries Survive?” The American Prospect, Nov/Dec, pp. 16-23.] Libraries might then “check out” these works to patrons as indicated in 7a above. Alternatively, if a licensing paradigm continues for access to online scientific and technical data, a portion of fees collected might be set aside or taxed to subsidize access to school libraries in rural and underserved communities [See Nunberg, 1998]. Development of standard licensing provisions and policies by libraries and publishers. [See, for example, the International Coalition of Library Consortia home page available online at <http://www.library.yale.edu/consortia/icocpr.htm>.] Development of university policies that mandate that professors and researchers must maintain full nonexclusive copyright/sui generis rights in any works or databases developed in their capacity as university professors or researchers. Establishment of a “collection society” for scientific and technical data (which has been proposed from year one, I think). MR. REICHMAN: Subpart d is a very important one. DR. SAXON: Let me ask whether this is a contractual matter that suggests how a group of persons might combine for a position of greater clout, as opposed to something that influences our recommendations with respect to a legislative regime? DR. BERRY: In another context it is called unionization. MR. REICHMAN: It doesn't apply to the legislative regime, but it does apply to defensive measures that science may take, such as a licensing consortium. It is all part of a licensing consortium, if you want. DR. BERRY: It is not the scientists. It is libraries and librarians. The ACS renegotiated its pricing policies for online journals as a result of having to meet with the consortium of libraries and librarians. DR. SAXON: So, you could take the hopeful conclusion that with good concerted effort there might be market adjustments to some of the things that are causing the greatest heartburn at the moment.
OCR for page 301
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS MR. REICHMAN: You might. The problem is that you have a sole-source provider problem. If there were some market pressures on the other side, then the ability of science to organize could influence those market factors. The anecdotal evidence from around the world that this database industry is highly dominated by sole-source providers. DR. RUMBLE: I think one of the interesting things is that the production of high-quality paper publications as, scientists have begun to expect, requires a great deal of capital investment. The Internet environment is changing that drastically. MR. REICHMAN: So, it may make them more broke. DR. RUMBLE: And I don't know one area of science where there aren't prestigious journals published by several publishers. DR. LOHR: I think it would take half the paper in Washington to list all the publishers of scientific material in the world. DR. BERRY: But there is a hierarchy. One thing that is probably about to happen is that as libraries cut back their subscriptions, the journals that get cut back first are the ones from the bottom of the quality list; journals that were generally available become less and less available. Now, maybe this will be filled in by electronic interlibrary mode, but I think it is going to cut the financial database. I would be willing to make a small bet that the number of specialized scientific journals is smaller in 5 years than it is now. DR. RUMBLE: I think the consolidation that has gone on is actually useful because a lot of journals, especially in Europe, have been on the basis of nationality—it's sort of like having a New York Journal of Physics and a Pennsylvania Journal of Physics, etc. Europe is now recognized as one entity, and that has been in some ways a very positive factor. MR. REICHMAN: Yesterday, we heard a presentation from the genomics data panel about a database created by drawing on many different databases. Each of those databases had a different set of rules, and some of those databases were private. The new database required input of elements from the private sector to be useful. So, the question is, What should the private person have to do? One possibility is discriminatory pricing or product differentiation. DR. BERRY: Individual versus library subscription prices, for example. MR. REICHMAN: Yes, individual versus library subscription prices, or could they give the data to the scientific community a little later and then charge less for it? These are the kinds of things that I think they do want you to think about. DR. SAXON: That is right, and there could be legal or policy recommendations. In particular, I think the key issue that you brought up, and I don 't know what the answer is, is what should the policy be toward transformative use? The ability to make transformative use is what we think is propelling scientific progress, and there ought to be an argument or a regime where the primary publisher provider should not be hurt. The power of what they are providing should be extended by the ability of others to make transformed databases. MR. REICHMAN: At least in the scientific world. Remember that in the commercial world, they are saying, “Hold it, if you make a transformative use of my database, I want a piece of the action. I want you to get a license from me, and then I want 10 percent of your profits.” I mean, that would be the equivalent of a derivative work in copyright law. DR. SAXON: Let us get the universities out of this discussion for a second. If you have one commercial provider and another commercial provider, you would most definitely in that regime negotiate a profit sharing. MR. REICHMAN: Yes, copyright law gives you the exclusive right to prepare derivative works of the expressive contents. We have taken the view that there should be no
OCR for page 302
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS derivative-work right at all with regard to scientists in the database. The question is, what should be the appropriate derivative database right with regard to commercial transformative uses? The problem is that everybody is a pig. The commercial second comers want a free transformative use, and the original creators want a totally controlled first right. In negotiations we argued that a compulsory license should be involved, that you should always be able to use the database for commercial purposes upon paying a reasonable royalty. DR. STEFIK: I raise a question about that issue, as well as Question 7, regarding the legal regimes you are talking about, because it seems to me that in this case of derivative works, it could get very complicated. It is not clear that one size fits all. I am thinking of this case of what is a transformational database. If one database is about A, someone has to address how A relates to B and then how B relates to C, and the relationship between B and F, and who contributed. There are all kinds of interesting questions about how that— DR. SAXON: Isn't there the other issue of what A really provided was the raw data and the copyrightable interface and way of accessing it and which piece of it was being utilized in a transformative way? DR. STEFIK: The best way to handle transformative uses is licensing, because nowhere else can you provide for the complexities of the individual cases. That is how you get away from that. DR. RUMBLE: There is an interesting difference between the way we handle two types of creativity, and intellectual property rights are associated with it. Under patents, we encourage creativity, and we want you to make an industry from it, but we limit the amount of time to much less than a person's lifetime. Take Prozac, for example. The patent on Prozac is running out, and during the time where that creator has had the sole market, other people have been able to tap into it and try to determine what to do with it when that intellectual property runs out. I think many of our very mature industries have grown on that leapfrogging, but under copyright and the new sui generis regime, we are imposing time frames for ownership that far outlast any human being. We are at the beginning of the information revolution not at the end of it, and if we stifle that intellectual creativity by saying that the primary database owner owns a right to a piece of the action forever, that is, I think in the long run, going to stifle creativity. MR. REICHMAN: You are saying that you have to be very careful. You are saying is that this is a new regime where data weren't protected before. Therefore, if you are going to protect the databases, the derivative product right would come last. It came last with copyright. Dr. Rumble is also saying that you certainly want to worry about the duration of any protection that you give. This protection should be shorter rather than longer. Unfortunately the European Union has skewed this issue of protection by picking 15 years out of the hat, and making it very easy to extend beyond that term of protection on the basis of posterior investments. Mark Stefik was saying something that is so pivotal; that is, suppose there is no right? Everybody is going to protect their work by licensing. I will put on my publisher's hat. A case can be made definitely for relying on licensing. We don't know what the information in the world brings. So, let us do it license by license. In order to do that we need a law that validates these licenses. MS. SINGER: I was sitting at the workshop yesterday thinking about Justin Hughes' comment that basically ISI's databases may not be covered by copyright, and we have been ambiguous about how they are covered. So, what did we do? We went into licensing big time. Every contract that ISI does with a consortium takes an enormous amount of time and we are
OCR for page 303
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS very, very fastidious in protecting ourselves. There were workshop participants in the back of the room who seem to ask why we need copyright, since we have licensing. I was thinking to myself that they are absolutely right because we have crafted now something much better than generic copyright. But from my perspective, because we are a global company, licensing may be fine. MR. REICHMAN: Let me pose this question for your discussion. Take fair use. You are all happy with fair use in copyright. Suppose the online contract says, “In order to get access to this database, you waive your fair use. This is a copyrightable database. Do you agree or not agree?” You cannot get on unless you agree. The proposed UCITA would uphold that. The Perlman-Reichman proposed amendment would subject standard-form, click-on licenses to certain restrictions —to a “public interest unconscionability” rule—on terms or conditions that disrupt the federal scheme. DR. BERRY: Jerry, I think one thing that you assume is in the traditional paper-publishing regime, the publisher issues the license. Now, in the electronic regime, I assign the license. I, as the data generator, should license you to publish for me. I shouldn't take a license from you for my data to be able to go back and use it over again. MR. REICHMAN: You are dealing with data generated by the scientific community. The concern here is that the scientist may need five chunks for a new database, some of which are from private databases. DR. BERRY: And if I want to do that, I should be able to go either to the database licenser and get a sublicense or go directly to the primary owner of the information and get my own license parallel to the database creator. MR. REICHMAN: You can do that. DR. BERRY: Only if I haven't signed away everything to you. DR. SAXON: In the real world we are talking about taking the secondary database from the primary database. The data that you want from the primary database may already be the aggregation of many different people's works. DR. BERRY: But the change is that I previously did not have the means to be a distributor of the data, and now I do have it with the Internet. DR. SAXON: I don't know if that is true. You may think that is true. MS. SINGER: That is somewhat true, I will say that. But you cannot imagine the millions of dollars that have gone into Chemical Abstracts' database. DR. BERRY: I completely agree, and that is great. MS. SINGER: Forget about the added value to the content. There also is added value to distribute that content so that you keep getting it in the correct archives with the correct search engines and what have you. DR. BERRY: That is also an important value, and I don't object to that at all. I just want to clarify that the potential role of the primary data generator in self-publishing has changed. DR. SAXON: The essence of this discussion is that self-publishing changes the regime. DR. LOHR: This is a stimulating conversation, but in a sense the future is very clear. The future is right here, right now. Because of a lack of action on the part of various parties, we are going to live in an intensely more highly licensed world in the future than we have even now. Already all of us who are in this business have erected these enormous edifices of licensing arrangements and we have done this out of our perceived ability to do so. Maybe that is wrong, but that is what has happened. The thought is that copyright, to the extent it even exists, is inadequate to protect what we feel are our rights, and so we will do these other things.
OCR for page 304
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS MR. REICHMAN: Two questions are driving this legislation. One is the existence of privately legislated intellectual property rights, i.e., case-by-case contractual licenses. The other question is, What would be the relationship between these privately legislated intellectual property rights and public intellectual property rights? If you get a right from Congress on which you can build your private regime, you have a super monopolistic power. A database protection right thus potentially expands your contractual powers beyond existing regulatory restraints. On the other hand, if you get something from Congress that says that you cannot do certain things with this right, and then you try to do it with contracts, that raises the question of what are the limits of contracts? So, a nexus is if you think that the scientists need free transformative uses, you need to preserve that transformative use regime, and this is a norm of science that ought to be ensured by the database law, which we have been arguing for. Then it is possible to argue in the context of the proposed UCITA that a standard-form license imposing a change from the free transformative use for scientific purposes violates a public-interest unconscionability rule (if one were adopted). Harvey Perlman calls it “impermissibility. ” They are now talking about fundamental public policies, but we are fighting about these limits. This is important because if you create a legislative pro-science policy it makes it easier to say that certain types of contractual constrictions are against public policy, and should not be enforced. DR. STEFIK: I agree, but I want to raise the issue of evolution. There is a sense that if you have a contribution, which is part of a larger database, and it is then used for something else, we have either no mechanism to simplify licenses or no mechanism in the legislative structure that crosses some threshold after which you are not going to care anymore. Yet if this were the music industry, it is really difficult and very expensive to track down all the infringements. In effect, you just don't even do that. MR. REICHMAN: There are two sides of this multimedia problem that is, from society's point of view, there is a lost opportunity. You are faced with so many licenses and such high transaction costs you won't do a project. That happens every day in the entertainment world because in order to do a multimedia project, they have to get 50 to 100 clearances, and any one can become a holdout. DR. BERRY: Groups perform their own music only. MR. REICHMAN: That is right. So, imagine what could become the problem for science. This is what we are trying to avoid is this multimedia syndrome. DR. STEFIK: One of the reasons I have favored licenses is because they didn't have all these complexities. Also, there was a kind of a faith that the licenses were just temporary. DR. RUMBLE: The reason I think licenses won't become simpler is because of the transformative use. People realize that there is money to be made just owning the rights to whatever data they have, but they feel more money can be made on that particular property that they own; its the Prozacs or Zantacs of the world that people are, I think, really holding out for. DR. BERRY: We have discussed the perspective of the scientific community and the community of the science world primarily from the perspective of profit making. But I think it is still true, I hope it is still true, that most scientific research in not-for-profit organizations is not motivated by a direct profit motive—that most scientists expect no immediate profit from what they do. The most important thing for scientists is to have their ideas disseminated and the most satisfying thing to them is not to get lots of money, but to influence the thinking of other people. Whatever recompense they get comes secondarily and tertiarily from the consequences of that. You don't win a Nobel Prize for a patent. You win a Nobel Prize for influencing the thinking of other people, and in a way this makes the ground rules for the users and data generators at the
OCR for page 305
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS primary level very different from the ground rules of the publishing community and the pharmaceutical houses and the end users that we have been focusing on. MS. WILLIAMS: But you have to recognize that in not-for-profit situations like universities and some of the research institutions when you know there is a marketable product, people will often quit that job and further the development elsewhere. DR. BERRY: There is much more of that now than there was 30 years ago. I still don't know a single astronomer who has patented a star. DR. STEFIK: Let me touch on another issue, which is this “first-sale” doctrine in electronic environments where there is no more than one user to use the work at a time. A first-sale person traditionally buys a hard copy. Let me point how we might legislate or have technological protection on first sale. I would publish and put something up for sale, and you are a purchaser. I would offer you two deals. One deal says that you can have first sale, which means you can get your copy and give it away. I am going to charge you $10 for that one. On the other hand, if you are willing to say that you are not going to have the right to give it away, I am going to charge you $3 for it. MR. REICHMAN: That is the discriminatory pricing and product differentiation mechanisms, which, in theory, ought to work to the advantage of science by allowing parallel new systems to be set up and special prices for science and related activities. DR. STEFIK: My point is that this is a first-sale doctrine. I am not sure which legal regime you are operating in here. If we are talking about individual contracts, there is no impediment to offering those two kinds of deals. DR. SAXON: And it is enforced by technology. DR. STEFIK: It is enforced by technology. If we are going to do these activities in some sort of legal regime, then I am going to build in barriers to that kind of marketplace action where you can charge differentially. MR. REICHMAN: This is a wonderful question, but there are no answers to it. One thing you should bear in mind as you think about the answer to this is that when you talk about contracts you have to be very careful. Again, just like fair use, contracts, as I teach it to my students, are based on recent methodological techniques for ferreting out true assent. As it comes up in this modern information environment, contracts increasingly are non-assent-based contracts. They are standardized, or contracts of adhesion. The terms are set solely by the licensor. This is sanctified in the proposed UCITA. That is very different from assent-based contracts of our tradition. Assent is not the basis of it. DR. LOHR: Just from our experience at Chemical Abstracts, and Leslie Singer might have a comment on this, these licenses are far from standard, and, in fact, to say that they are non-negotiable is just naive. We haggle for years on some of these licenses. This idea that people who are in the business of providing scientific databases are monolithic monsters that can impose whatever they will on the world just defies reality. MR. REICHMAN: I disagree with you. Of course, you do negotiate, and if we get to negotiate contracts we are not at all afraid of the outcome. The negotiations between the parties in these complex contracts will actually reveal what are the empirical realities of the information world, but that is not what is driving the revision to the UCITA, Dr. Lohr. The revision of Article 2B is driven by several very large companies that are adamant in wanting to be able to validate standard-form contracts that will be imposed in Internet delivery and in which you only have the choice to agree or not agree to the terms as stated in click-on licenses. Those contracts at the moment are valid only in one circuit, the U.S. Court of Appeals for the Seventh Circuit.
OCR for page 306
PROCEEDINGS OF THE WORKSHOP ON PROMOTING ACCESS TO SCIENTIFIC AND TECHNICAL DATA FOR THE PUBLIC INTEREST: AN ASSESSMENT OF POLICY OPTIONS Most other courts have declined to uphold those contracts. The proposed UCITA (formerly UCC Article 2B) as it is presently written, says that these contracts are uniformly valid, and you have no recourse. DR. SAXON: What we have here is another case where a matter which is supposed to remedy or address a situation in one arena possibly could be interpreted as providing unintended consequences in the scientific arena. MR. REICHMAN: I am not suggesting that most scientific database providers are going to behave unreasonably. What I am suggesting is that some large scientific publishers will behave anticompetitively if they are given the license to do so. DR. SAXON: Big databases for which there is a massive amount of negotiation are different than a particular item where the individual cost is very small. They really are very different situations. MS. WILLIAMS: I'm afraid that we are out of time. I would like to take the opportunity to thank our panelists and participants in the session.
Representative terms from entire chapter: