For Community Remote Sensing
OpenStreetMap of Tacoma, Washington, done by crowdsourcing
OpenStreetMap of Washington, D.C.
External Relations and Legal Officer
EDITOR’S NOTE: See story about the
report on CRS from the United Nations’ Space-based Information for
Disaster Management and Emergency Response (UN-SPIDER). Also,
LBx Journal’s Spring issue includes a major report on Location Data
Privacy, in conjunction with The Location Forum
Community remote sensing (CRS) is becoming a powerful tool with many important applications. CRS means that the community of users contributes data points to certain projects, from the field, from mobile devices, to a centralized digital map serving as a data aggregator. Similar concepts are crowdsourcing (more casual) and VGI (Volunteered Geographic Information).
CRS has the particularity of using multiple types of location data and sources. Analyzing and combining all or some of these different types of data create a new set of data from which information can be extracted, leading to new knowledge. It is the fact of combining all this information that creates a more accurate picture of the situation at hand, whether it is for humanitarian response after a natural disaster, for a scientific study of an ecosystem, for an applied use such as microweather predictions in mountainous country, or for many other uses.
As with every human activity, legal issues need to be taken into consideration. At each stage of a CRS project – data collection, data access and usage, sharing and distribution of data and/or information – some legal threats can be found and it is important to be aware of them before embarking on a CRS project.
Four main legal concerns (first identified by Kevin Pomfret, attorney at the Centre for Spatial Law and Policy) are national security, privacy, intellectual property rights, and liability. Failure to address these legal issues when considering a CRS initiative may eventually harm the project or have negative impacts on future projects.
National Security Considerations
A State’s national security policy might restrict access to specific data, such as very high resolution satellite imagery, or to aerial imagery of its territory. While this should always be taken into account by CRS projects, and can certainly be a limitation for some, the impact of these policies remains quite
limited. In fact, International Law, as well as the United Nations
system, has acted in favor of access to data, through the 1986 United
Nations Principles relating to remote sensing of the Earth from space,
and through the implementation of the International Charter on Space and
Much more pressing are the issues of privacy, intellectual property rights (IPR) and liability.
Privacy is a complex notion. It has different meanings in different cultures and enjoys different levels of protection according to State law, as each State is sovereign with respect to how it applies the right to privacy on its territory.
National legislation mostly protects against violations of privacy by public authorities, but no mention is made of private or commercial intrusions. These cases then need to be dealt with on a case-by-case basis, which does not offer an environment of legal certainty. Interestingly, cases such as privacy infringements by Google Street View have been dealt with differently in different European States. While Germany declared Google Street View as legal, Swiss courts ruled that Google must take action to protect privacy by blurring out faces and license plate numbers. In some countries, the term “privacy” does not even translate directly and is not protected as a right.
Intellectual Property Rights Virtually everywhere, a person producing any kind of printed, written or in some other way publicized image, data or text is protected by copyright, by default. This means that the person who created the data needs to receive at least proper attribution when the work is re-used. Copyright is, however, not always the most appropriate way to protect remote sensing data, as it cannot cover raw data, which is not an original creation, but only processed data. Other forms of legal protection may apply to raw data, like database protection, or ownership rights.
Policies are usually in place for using remote sensing data, and in most cases a license needs to be acquired. The license details the conditions under which the data can be used, which can be more or less restrictive. A few typical conditions include:
- the data shall be used only for the particular purpose for which the license is granted,
- it shall not be used for any purpose that would be against the law,
- it shall not be altered,
- it shall not be further distributed.
These last two conditions are usually not compatible with CRS projects, as the aim is usually to add information layers and then further distribute the new data. Very often, there is no particular licensing system in place, and this may result in even more confusion. This does not mean the data are not protected; it just means it is more difficult to know which type of usage is allowed.
Many organizations providing data free of charge, such as Google, do have a policy and terms and conditions for use of their data. However, as each organization may have a slightly different policy, it becomes quite difficult to keep track of all of this in a CRS project, where there are many layers of data. Therefore it is more prudent for a CRS project to rely only on original data created by the contributors or on data in the public domain. This limitation, however, drastically reduces the amount of data that can be used. A good example of a project that has adopted this angle is OpenStreetMap (OSM). See Figures 1-2. The OSM copyright and license policy, as well as the legal dispositions, specify: “Only sources with compatible licenses – such as U.S.
Government information released into the public domain – may be used as bases for adding OSM data. However, it is OK to use Yahoo! Aerial Imagery, as Yahoo! has agreed to allow OSM to use it. Better still, create the data yourself!”
This is a good example of how to avoid legal issues in a CRS project, but it also provides evidence that a legal system with a clear and uniform approach to licenses and with a better overall comprehension of the uses of technology could make such initiatives much easier to handle and also more efficient. Another complex question concerns the term “derived work” and what it covers. If only a very small part of the data is used to create a new set of data, should the licensing policy of this data apply to the whole compilation? As CRS is usually a compilation of many different data, it can be very complex to attribute a coherent legal status to the compilation.
The solution found by Open-StreetMap is therefore one of the safest ones. All the data used is in the public domain, or specifically created for the project – the contributors having previously agreed to the OSM terms – and the compilation is protected under a Creative Commons license. Initiatives such as Creative Commons are very useful for CRS projects, as they answer their need for a standardized, internationally applicable and legally acceptable way of distributing and sharing the data.
Liability in the case of CRS is directly linked to data quality. How can the quality and the accuracy of the data be trusted and can it/should it be verified? This is one of the limitations of CRS and of all participatory or “open” projects, like Wikipedia, for example. The question is, will this liability issue limit the use of CRS data or harm the development of this type of initiative? Whereas it is unlikely that it will harm the development of CRS, the lack of guarantee of data quality might very well limit the uses made of CRS data for critical issues like disaster response or science model building. As noted before, using and sharing are the main purposes of CRS data, and limitations to them should be avoided.
If data can never be 100% guaranteed, still there are levels of trustworthiness, depending on who the contributors are, how and if they are selected, how many people are involved and how the project is managed. Something as wide open as Wikipedia cannot be compared to the GISCorps project, where a small number of professional volunteers are carefully selected for each project and where one volunteer is a designated manager.
Two examples of cases in which uncertainty over data quality could harm the use of CRS data are political pressure and lawsuits. In the courtroom as well as in the political arena, the only acceptable arguments are ones that are verified, verifiable or coming from reliable sources.
It would be counterproductive to force organizations involved with CRS to guarantee their data and to assume liability for them. This would probably be the death of most CRS projects. Another approach would be to standardize the use of liability disclaimers for CRS projects. This aspect goes hand-in-hand with the technical challenge of metadata requirements in CRS, and should probably be tackled simultaneously, through cooperation between technical and legal experts in the field.
Laws are sometimes uncertain and outdated and usually do not reflect the level that technology and its applications have reached. This is not surprising, as laws are usually reactive and answer a need after a new situation arises. One of the most pressing issues for CRS projects seems to be related to standardization and a common understanding of the rules and the technologies involved. Concerning privacy, the way forward will probably go through custom rather than through hard legislation. As people grow accustomed to new technologies, their views on the boundaries of privacy change. Over time, cultures change, and in this technological age, cultures seem to change ever more quickly. It is unrealistic for a system of law to keep pace with these changes, and it will probably evolve slowly as an adaptation to a fluid reality.
Concerning IPR, without harmonizing the laws worldwide, there could be an effort made to raise awareness about community remote sensing and its goals, which are mainly scientific, educational and related to disaster management and other public good. If specific licenses could be designed for this kind of project and on a global basis, use and sharing of the necessary data might be facilitated. Efforts have already been made in this direction, such as the Creative Commons initiative.
Finally, several international institutions are already working on the different issues at stake. The Data Democracy Initiative of the Committee on Earth Observation Satellites (CEOS) is particularly worth noting. This initiative aims at providing “timely access to key datasets free of charge to build capacity worldwide, especially with respect to developing countries. Additional Data Democracy initiatives include enhanced data dissemination capabilities, sharing of software tools, increased training, and technology transfer to end users. CEOS member agencies recognize that the GEOSS Data Sharing Principles should serve as the basis for data access in this context to contribute data for the public good. In particular, CEOS agencies will contribute to the GEOSS Data Core by making several datasets available on a full and open basis.” (Source: CEOS Rio Statement).
The CRS community, as well as all citizens and businesses involved in social networking and crowdsourcing, will probably show the way. As CRS flourishes, a few broad and common principles for these issues should be found and agreed upon internationally.