Examining the Feasibility of Adopting the Protocol for Implementing Open Access Data for Database Products of the United Nations

STATUS: Ready for review, although revision is a continuous process.

In the course of the work of the United Nations (UN), it frequently engages in partnerships with private or non-private entities leading to the production of databases ( see and examples). The production of these databases or other intellectual property works requires funding, and the use of this database is often necessary in order to pursue the project for which the database was created, whether by the UN in continued partnership with the third party entity, or by itself. The UN therefore needs to enter into licensing agreements calling for the grant of licenses (ideally a perpetual license), allowing for the non-commercial use of the database. The negotiation of these licenses is necessary in order that the UN can continue to use this data without having to pay for it, but this process is also complex, time-consuming, and a drain on UN resources which could otherwise be used in areas UN legal resources may be more needed. An idea that could be considered by the UN is to place the data under the Protocol for Implementing Open Access Data created by Science Commons which seeks to enhance the sharing of information in databases, calling for the placing of database in the public domain.

Placing the data under the Protocol can serve not only to free up UN legal resources from negotiations of license agreements but can also enhance the distribution of data produced by the UN. The Protocol was adopted precisely in order to make the sharing of data easier for researchers. In the United States for example, a database constitutes a "compilation" and is subject to copyright protection if the work is formed by the collection and assembly of data that is selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of author. The rule is slightly different in China where there is copyright protection for databases if there is originality in the selection and arrangement of content. In Canada, to receive copyright protection, the selection and arrangement of the components that make up the database must be the product of an author's exercise of skill and judgment, and in Europe where database producers have sui generis rights irrespective of whether the database is in itself innovative . These differences impede research, and worse, enable data providers to dictate the research that can be done using the data. Science Commons argues that imposing that kind of control threatens the very foundations of science, which is grounded in freedom of inquiry and freedom to publish. The Protocol addresses these issues by calling for certain standards in the distribution of information in databases.

The Protocol calls for a waiver on the relevant rights on data and an assertion that the database provider makes no claim on the data. Citation or attribution will not be based on legal requirements but on the norms of the scientific discipline. Second, the Protocol calls for the waiver of all intellectual property rights, including copyrights and sui generis rights. Third, the Protocol requires an affirmative declaration that contractual constraints do not apply to the database. Lastly, it calls for the provision of interoperability with databases and data not available under the Protocol, through metadata. Thus, the Protocol goes beyond merely placing the database in the public domain. It attempts to create a system where there is legal predictability, certainty, ease of use and comprehension, and the imposition of the lowest possible transaction costs on users.

However, certain aspects of the Protocol would be difficult for the UN to accept. The UN currently provides database information to the public through its UN Statistics Division under the UN Data website, and if we treat the Conditions of Use imposed by the UN Data as an indication of its philosophy of data sharing, then the UN will have problems complying fully with the Protocol. It is useful to treat the these restrictions in the Conditions of Use as restrictions that the UN will seek in sharing data because these same restrictions are found in the UN Conditions of Contract which are templates of contracts used by the UN in its agreements with third parties.

UN Data provides that all data and metadata on the UN Data website are available free of charge and may be copied freely, duplicated and further distributed provided that the data is not put up for sale or otherwise commercially exploited and that UN Data is cited as the reference. This goes against the Protocol's call for a waiver of all intellectual property rights to the database, and for citations to the database to be a matter left to scientific discipline rather than contractual requirement. There are at least two reasons which are apparent as to why the UN would not be willing to do this. First, the UN needs to maintain its relevance in as many aspects as possible, both in order to maintain its primary status as an important body in world politics and to justify the assessment of financial contributions on member states. Having data that it has aggregated attributed as the basis for economic studies and scientific discoveries is an important means of achieving this goal.

Second, the UN operates on extremely scarce financial resources and would want to be keep for itself the possibility of any financial advantage that can be taken on its research and work. The restriction provided in the UN Data’s Conditions of Contract would not be effective in preventing an entity in a jurisdiction where there is essentially no copyright in databases from formulating the data in a different manner and releasing the same under the Protocol. However, by placing the data under the Conditions of Use, the UN could sue under a breach of contract claim where the terms and conditions of the use are violated. (See Ticketmaster v. Ticket.com) Through this, the UN can have a means to try to cause the cease and desist of the commercial production.

The Protocol however seeks to accommodate as many cases of sharing of database information as possible, which would be useful for the UN as a replacement for the license agreements it enters into with private parties. Further, the UN Conditions of Use, that the data may only be used if the user agrees that the UN shall not be liable for any loss, damage, liability or expense incurred or suffered that is claimed to have resulted from the use of the UN Data, on any fault, error, omission, interruption or delay with respect thereto, is not inconsistent with the Protocol. Thinh Nguyen, Legal Counsel of Science Commons, says that disclaimers of warranties and limitations of liability do not conflict with the Protocol. “We feel that this represents just an attempt to allocate risk, not reduce research freedom, and we encourage data providers to consider making those disclaimers whenever possible. This may encourage more data providers to make data openly available, if they can reduce the risk of liability to themselves.” Further, the desire of the UN to avoid the differing legal obligations in the jurisdictions where it operates is solved by the simplicity of the Protocol, which calls for scholarly conventions in attribution rather than legal standards.

The adoption of the UN of the Protocol in the data it releases that is produced whether by itself or in collaboration with third parties will greatly simplify the means by which the UN manages its database. It is therefore necessary for the UN to resolve the issues that do not comply with the Protocol, so that it can take advantage of the mechanisms of the Protocol.

References

• Simon Chesterman, Law and Practice of the United Nations: Documentary and Commentary, Oxford University Press (2008).

• Thinh Nguyen, Freedom to Research: Keeping Scientific Data Open, Accessible, and Interoperable.

• Thinh Nguyen, February 15, 2010, Email to author, on file with author.

Comments of Professor Moglen on the first draft:

The adoption of the Protocol involves a deprecation of the Science Common's earlier position on that databases can be licensed for use under Creative Commons licenses. Previously, Science Commons advocated the use of Creative Commons licenses for databases, based on the fact that databases, or some aspects thereof, are subject to copyright. The license means therefore that certain rights granted under copyright are waived. But researchers who need to draw from databases find these restrictions difficult to work with. Some of their difficulties consist in dealing with differing and overlapping data sharing policies, and with agreements, laws and licenses which have conflicting obligations, limitations, and restrictions. These difficulties impede research, and worse, enable data providers to dictate the research that can be done using the data. Science Commons argues that imposing that kind of control threatens the very foundations of science, which is grounded in freedom of inquiry and freedom to publish. The Protocol addresses these issues by calling for certain standards in the distribution of information in databases.

• I think this account fails to take account of the fact that databases are not in fact subject to copyright under the law of the United States, which is the place of publication of a large proportion of the world's scientific information. So the Science Commons position that copyright should be used to protect freedom through CC licensing was subject to a more significant legal criticism than the administrative burdens you discuss. If this problem is to be discussed at the level of detail you're discussing it, clarity about the global legal treatment of data compilations is crucial.

The Conditions of Use of the UNdata provides that all data and metadata may be copied freely, duplicated and further distributed provided that they are not put up for sale or otherwise commercially exploited. UNdata should also be cited as the reference. This goes against the Protocol's call for a waiver of all intellectual property rights to the database, and for citations to the database to be a matter left to scientific discipline rather than contractual requirement. Several reasons are apparent as to why the UN would not be willing to do this. First, the UN needs to maintain its relevance in as many aspects as possible, both in order to maintain its primary status as an important body in world politics and to justify the assessment of financial contributions on member states. Having data that it has aggregated attributed as the basis for economic studies and scientific discoveries is an important means of achieving this goal. Further, the UN operates on extremely scarce financial resources and would want to be keep for itself the possibility of any financial advantage that can be taken on its research and work.

• Your second point misunderstands the situation. If the UN were to make data available under the Protocol, it would not be preventing itself from making commercial use of the data. Nor would it be precluded from offering the data for inclusion in commercial products at a fee. There would simply also be another copy of the data available under free terms. You should be more careful to consider the legal situation, and the economic possibilities of dual licensing.

Another divergence from the Protocol is that the Conditions of Use impose license requirements -- the data may only be used if the user agrees that the UN shall not be liable for any loss, damage, liability or expense incurred or suffered that is claimed to have resulted from the use of the UNdata, on any fault, error, omission, interruption or delay with respect thereto. This arises from the fear of the UN of any liability, that would further strain its cash-strapped operations.

• Interpreting the Protocol to prevent indemnification requirements for users of data is possible, but defeats the whole purpose of the Protocol, which is surely not to spread risk back onto data publishers. Everybody making data available under the Protocol has the same interest here, as does Science Commons. I have never been counsel to Science Commons, and I can't speak for it or about its legal positions, but I think your assumption here is at best captious and probably wrong.

The loss that arises from the UN's expected refusal to adopt the Protocol does not restrict anyone from being able to access the UNdata and other information and databases that the UN releases. But by not implementing the Protocol, an important provider of databases does not join in the Science Common's project of enhancing the interoperability of information systems. Science Commons does not appear to have launched the Protocol as yet. As an entity operating in multiple jurisdictions, the UN is in a unique position to aggregate database information. Its expected non-implementation of the Protocol will present important issues in the harmonization of policies on sharing of databases.

• Because the UN data is freely available, what's to prevent someone else from making it available under the Protocol? Anyone who needs it in that form will put it into that form, without any of the problems you discuss above. That justifies the UN in doing business on its own disclaimers, right? If not right, you need to show why.

• I should have said, "anyone in a sensible place (other than Europe)where there is essentially no copyright in databases can formulate the data however they want and release it under the Protocol if they want to, thus justifying the UN in doing business on its own disclaimers because they are essentially worthless: the UN and its agencies cannot effectively sue to enforce them and they make the data no less free than it was under relevant national law anyway." (email to author)