But where's the property right in the first place? Who owns what here?

Rules about harms to property are relevant where there is property. Because you haven't shown where property rights exist in this setting, I don't know how to understand the comparison being made.

-- MichaelRhodes - 21 Oct 2014

Property Models for Internet User Data Protection

In this essay, I will further analyze the topic in the course that has interested me the most so far, namely analyzing the negative externalities caused by social network data mining to third parties within the lens of property law. I will ignore contained social networks such as Facebook in which every member agrees to a privacy policy although most members likely do not read or understand the privacy policies that are put before them. I will focus instead on other social networks like gmail where members can be messaged and data mined outside of their network without even the preface of having agreed to surrender data. First I will try to explore the limitations of using traditional property negative externality analysis and after making these concessions I will look to the policy impacts of treating data mining as a negative property externality.

There are many similarities between social networks and traditional property communities. Both are built around communities, properties are generally designed into residential and commercial areas in order engender value creation among people where social networks are designed to connect people of similar values and interests. For example, on LinkedIn? I am able to join a group of intellectual property litigators. Additionally, networks and communities both create positive externalities from holding multiple members and would have little to no value if created on isolated islands. Finally, the positive externalities created in both communities are offset by negative externalities that contain free rider problems where the externality generator only faces a fraction of the cost of the externality.

There are differences between non-voluntary social networks and land property communities. One difference would be the concentration and magnitude of external nuisances. In a traditional community, nuisances including noise and pollution are diluted the further they are away so offenders could at most be a few miles away. Social networking nuisances are global in nature and thus one person can be dealing with thousands of global nuisances that all open the person's data to malicious use. On the other hand, these nuisances could be life threatening such as the encroachment of pollution or the threat of a nuclear power plant meltdown. I am not convinced that data mining by advertisers and the government would be as damaging to an individual's freedom as a nuclear meltdown.

Additionally, the costs of avoiding nuisances are different in a property and social networking context. For example, to avoid noise pollution a person may undergo the cost of moving their house or. This would undergo a large upfront cost but be cheap to the user after eating the cost of moving. In the gmail context a user would have to block off all senders from any data mining email address. This would cost the user only a few mouse clicks, but can come at the cost of lowering their ability to socially and professionally network. While past generations are not reliant on email and electronic communications, it is becoming increasingly expensive for young people to disconnect from the human informational exoskeleton. Despite the differences between traditional property nuisances and the negative externalities imposed by social networks I believe it could be illustrative to analyze social networks in a property context.

Negative externalities in a traditional property context are controlled in a few ways. Substantial and unreasonable nuisances are regulated by courts. One exception to this liability is when the nuisance is a usual and known use of that land. For example, if you purchase a property next to an already established garbage company you cannot complain about the smell. One could argue that loss of privacy is usual and known since data mining has become so widespread on the internet, but I believe that almost all consumers are ignorant of the extent of data mining that goes on behind the scenes so this is not a satisfying answer.

Courts apply a balancing test to see whether to apply injunctive relief or money damages. The four part test weighs the likelihood of irreparable harm, if a remedy at law exists, the balance of harms between the parties and the public interest. In this context we will weigh the interest of internet consumers having informational privacy against the interest of social network providers like gmail to sell information in order to fund the services that they offer for free. One could argue that dispersing a client's information has a high likelihood of irreparable harm since the information cannot be put back into the bottle after release and government agencies could potentially use the information to deprive a citizen of their liberty. On the other hand, the vast majority of consumers will face no tangible harm from the disbursement of their information and one could argue that a criminal of the state has no right to freedom just because their crimes are private. Currently no satisfying remedy at law exists for social network users although some lawyers are trying to develop one . Preventing social network providers from data mining would harm them by removing one of their largest revenue sources and costing them billions of dollars. The analysis becomes more challenging as to the public interest since there are lobbyists and zealots on both sides of the freedom of internet consumer information.

In conclusion, I believe that a balancing test would weigh for the protection of internet user data if this issue is analyzed as a property nuisance due to the lack of an available legal remedy and how the harm to providers would only be a transfer of value from users. Personally, I believe that data mining is likely to create irreparable harm and that public policy favors the protection of data but these test elements are not needed.

<--/commentPlugin-->