Interview on Obscurity
for GOOD magazine 2016
DJ: What were the origins of this project? That is, how did you first come across mugshot websites, and when exactly did you want to make an art and social justice project out of them?
PC: When I came across the mugshot industry I was just shocked that it could be so spread and legal. I’ve always been researching on the Right to Be Forgotten, and I was even more shocked when I started to find op-eds in major American newspapers trying to demonize it, and suggesting that the Europeans have a law trying to censor the Internet, which of course it’s not the case and it is an argument used to not let American citizens exercise their basic rights and keep the giant Internet companies exploiting personal information.
By being a European living in U.S., I notice directly huge differences between the regulatory system in Europe and the total lack of legal protection for consumers in United States. This is the case also for the welfare and the criminal justice systems in U.S. that eventually created this horrible situation of mass incarceration of poor people that then even get shamed on the Internet for the rest of their life.
DJ: Why is this type of "information abuse" so wrong?
PC: It’s so wrong because in most of the cases the individuals in the mugshots don’t have the resources to take down their humiliating pictures, they can’t afford lawyers, nor paying the extortionary fee to remove the mugshot, or they might not even have access to Internet because they are in prison. The mugshots circulate online before of any trial and therefore many individuals are exposed even if they are innocent. And even if there are laws and resources that can be used to take legal action for taking down the mugshots, it might be not enough, because the owners and the servers of mugshot websites are often in other countries and anonymous, thus shielded legally. That’s why it really comes down to the responsibility of search engines to remove that kind of sensitive information like mugshots.
In the past years I’ve been working several projects concerning the circulation and exposure of sensitive information over the Internet. In particular, I made artworks with recontextualizing photos of individuals on Facebook 1, Twitter 2, Google Street View 3, and other online platforms 4.
These provocative works were meant to warn about social media companies’ privacy policies and the danger of losing control over their pictures posted online. However, in those online projects I never let the search engines indexing that personal information, not exposing their names. By just using the robot.txt I kept away the crawlers. The data was contextualized in an online artwork, where everyone could be aware of the artistic experiment in which those pictures were used.
With Obscurity, instead, I try to boost the indexing by the search engines of the obfuscated criminal data as a strategy to make it harder to identify individuals online. Creating noise in such of big data can be considered a form of hacking, by intervening in complex systems of distribution, management and control of information, from the technical to the legal spheres.
DJ: How exactly did you clone the websites?
PC: I use very similar domain names of the original mugshot websites, then I copied their logo and CSS layout and I made a very similar interface. For months I had scrapers harvesting mugshots and data from four original mugshot websites and save that huge amount of data on a secret server of mine. While the cloned websites are published through a proxy public server to protect the original data I harvested from attacks of any sort. All the cloned websites can eventually multiply and be mirrored very quickly if a mugshot website tries to take down the project.
DJ: How does your algorithm blur the images?
PC: It’s a relatively simple process made by a few lines of code. It resizes the mugshot to small PNG files and then it apply for more than hundred times an image filter available with PHP 5.5. The final effect is very interesting, as uncontrollable. It took me several experiments to feel comfortable with the amount of blurriness and also the visual impact that the pictures could provide. The computational power to process 15 million of images is unaffordable for me, and so they are made on the fly when the individual profile is opened on the browser. It’ll take some months to have all the mugshots processed. Big data is often something that human beings can’t really see in any case, yet each record can have a huge impact on a life of an individual.
DJ: How does the algorithm shuffle data on the people in the mugshots?
PC: It finds the original data from the database that I assembled, then for each page of results it sort profiles for gender, race, and age, and shuffle their name for each of those categories. Therefore in the page of results, each individual profile corresponds to someone with same location, gender, race and age, but with the picture and names of others in the same category. That also means that in the same page of results there is the original name and mugshot, but it’s impossible to identify because it’s shuffled with a dozen of others. Instead, in the profile page of the arrested individual, all the data is maintained accurate, like color of eyes, and especially the criminal charges, yet the names and mugshots aren’t the real ones. That’s to make public the data about the arrest without exposing the person arrested.
DJ: How did you boost page rankings so your clones are promoted over the original mugshot websites?
PC: By having a similar domain names and other techniques in the html code. When someone is looking online for a specific names of someone plus the keyword “mugshot”, the search engines will provide the page and the picture of someone very similar, but not actually identifiable. The more this project became popular and get linked on news articles, like this one now, the best ranking the cloned websites will have, while the original ones will tends to be pushed out from the search engines results.
Can you explain how the cloned sites’ visitors can participate in the project by judging an arrested individual and deciding to keep or remove their data from the mugshot websites?
I wanted to include the participatory element to infer about how the process of making sensitive information available should be publicly discussed. The way the information flow is managed in our age is really at the core of our democracy. It should’t be controlled only by companies and authorities, there should be a democratic process to let people decide what should be public or private, accessible freely or not.
My proposal is extreme, pushing boundaries with art, as eventually the democratic process becomes a popular jury open to everyone, where the people can judge to condemn or give mercy to who has been arrested, with making public their information or removing it. We might see this form of online judiciary in the future, and many forms of online shaming are already happening to expose people with bad social behaviors, it happens any instant on Facebook and Twitter, or for instance with rating AirBnB’s hosts and guests and so on. We will need to learn how to judge people online civilly. It might take hundreds of year to build such society.
DJ: How does this project intersect with Right to Be Forgotten law proposals?
PC: It proposes a form of Right to Be Forgotten that can be workable for the U.S. My proposal with the campaign Right to Remove is a simplified version, suggesting very basic categories of personal information that people should have the right to remove from search engines results, for instance, information about minors, sexual orientation, and civil courts records. For that I launched a petition to be addressed to the Federal Trade Commission and the Congress. It should be a nationwide bill to give this legal right, not something like now, with a very fragmented legislation, different in each state, which obviously doesn’t make sense because the Internet is not something federal. For instance at the moment only in California there is a law that lets minors to erase the information they published till eighteen. This is a proof of how bad privacy rights are in U.S. in 2016.
The Right to Be Forgotten is not perfect, however in U.S. there is the need of basic rights which can’t be delayed because of Google lobbying against it or legislators debating on a convoluted and outdated legal code and system. Obviously, information of when you were minors or your sex life don’t need to be searchable online as many other types personal information indicated in the Right to Remove campaign.
DJ: What is your ultimate hope for this project?
PC: Bringing attention to the problem and making people aware of the need of a form Right to Be Forgotten that works and it’s fair.
| | | | | | | |