A new empirical research paper I have coauthored with Catherine Tucker of MIT-Sloan examines the question of how Edward Snowden’s surveillance revelations have shifted the way people search for information on the Internet. We look at Google searches in the US and its top ten trading partners during 2013. We identify a roughly 5% drop in search volume on privacy-sensitive terms. In the US, UK and Canada, the countries in our data who were most involved with the surveillance controversy, search volume fell for search terms likely to get you in trouble with the government (“pipe bomb”, “anthrax” etc.), and for searches that were personally sensitive (“viagra”, “gender reassignment”, etc.). In France and Saudi Arabia, search volume fell only for the government-sensitive search terms. This paper, though at an early stage, provides the first systematic empirical evidence of a chilling effect on people’s search behaviors that is attributable to increased awareness of government surveillance. I will be presenting this paper at the Privacy Law Scholars’ Conference in DC in June, 2014. I would welcome comments at email@example.com.
We heartily recommend that you read the whole thing, but here are the striking take-aways.
We are spending 6% more on incarceration than we are spending on education. Low-level drug offenders sentenced under mandatory minimum laws are driving a substantial portion of the costs. Offenders are routinely overclassified into higher and more expensive levels of security than they really need. And Massachusetts is not being strategic about its incarceration spending to make sure that it is getting the least reoffending for a given budget.
A very human desire to lock everybody up for ever takes no account of the costs, or of how doing that crowds out investment in other things we might like more of, like better education, lead abatement or public transportation, which also in turn have a positive effect on crime down the road.
I wonder if the Lege and the AG’s Office are listening? Or will they keep wasting our taxes on strategies that don’t work?
[VIA: Sentencing Law and Policy]
Science Daily reports that one of the technological limitations on cellphones’ tracking capabilities is about to be lifted.
A research team led by Professor Dong-Soo Han of the Department of Computer Science at the Korea Advanced Institute of Science and Technology has developed a way of locating cellphones using their WiFi fingerprints to within 10 meters in indoor locations in cities.
The article is pretty gung-ho about the capabilities of this new technology, though they also report Professor Han as suggesting that “There seems to be many issues like privacy protection that has [sic] to be cleared away before commercializing this technology.”
Yes, Professor Han. That may be something of a concern.
In particular, may I ask, pretty please, that legislators considering bills to protect the privacy of cellphone users’ location data, bear in mind that this kind of indoor location detection has been shown to be possible? I give law enforcement oh, about five minutes after the commercial release of this technology before they start using it in investigations.
The obvious Fourth Amendment issue here is that the current rulings allowing law enforcement use of cellphone data tend to rely heavily on the fact that when you are outdoors, you are not generally held to have a reasonable expectation of privacy, and that your location is reasonably observable to members of the public. If cellphones become capable of tracking you indoors, where you do typically have a reasonable expectation of privacy, then judges will have to choose whether to weaken the Fourth Amendment further by not requiring a warrant for the seizure of cellphone data, even though it contains data that would be considered private, or to strengthen the Fourth Amendment in the light of cellphones’ increased capabilities by barring its use without a warrant. These are rough waters for any jurist, which is part of why we strongly support cellphone location privacy laws that unambiguously require law enforcement to obtain a warrant before being allowed to collect cellphone location data.
The research is reported as Hyunil Yang, Giwan Yoon, and Dongsoo Han, “Floor Accuracy Improvement of Wireless LAN based Large Scale Indoor Positioning”, IEEE MTT-S IMWS-IRFPT 2011, KAIST, Daejeon Korea, p.89-90 (2011).