Page 68 - AC/E Digital Culture Annual Report

68Inside the PanopticonThe constant production of large amounts of data in real time through the social media also has a sinister counterpart. It is not unusual for Big Data to be compared to Big Brother or, better still, to the Panopticon – a type of peni- tentiary building devised by Jeremy Benthan in the eighteenth century which creates the sensa- tion of being constantly watched – especially in the wake of the Edward Snowden case. Govern- ments monitor citizens to ensure their security; this is by no means new and is part of the history of power structures studied by Michel Foucault, among others. In the modern state people are watched and, at the same time, encouraged to reveal their deepest secrets through confession, psychoanalytical therapy or, nowadays, by posting their “statuses” on Facebook.As we have seen, the object of study of the humanities tends to be external, autonomous and nished – a historical document, a literary text, a visual representation – and research therefore does not usually pose ethical dilemmas on the privacy of creators and recipients. However, as consumers of culture, our acts are registered every time we search for a book, lm or song on the internet, and when we click ona product and buy it; the same is true when we visit a museum – the surveillance camera is there to protect our heritage from crime and theft, but also to keep check of visitors; lastly, when we borrow a book from a public library a record is created in the database.We should ask ourselves how humanists can study citizens’ cultural habits, in constant dialogue with libraries and museums and using methods to anonymise data.The case of public libraries is particularly inter- esting because they are a type of neighbourhood infrastructure accessible to everyone regardless of their economic status. Librarians record all loans, noting the date and borrower, in their databases.Nevertheless, this type of data is not accessible because municipal libraries have a long tradition of data protection (Starr, 2004). They do, how- ever, publish lists of the most frequently borrowed books which function as indicators of contempo- rary taste. In order to be studied, this data would have to be published in an open format like XML or CSV and include a series of metadata such as the place and time of the loan, but such practices would encroach on users’ privacy.For researchers interested in reading habits, enjoying access to so much data would be a breakthrough. For example, it would be possible to ascertain how lms, television and advertising in uence people’s tastes and reading habits. Manufacturers of electronic books, for example, are already using reading statistics to discover which books can be regarded as good – because readers nish them – despite not being best sellers; or to identify the next Dan Brown based on readers’ degree of satisfaction with books written by unknown authors (Kobo, 2014). Ba- sically, all the data generated by our electronic books is amassed by publishing companies to learn more about the relationship betweensales and customer satisfaction; this makes it easier to justify economic decisions about the publishing future of a particular author, literary saga or genre.By this I do not mean to imply that public libraries and museums should act in the same way as companies. I merely wish to point out that the state of being watched existed before the social media – just as spaces of resistance did. Just as companies like Twitter have been accused of exerting coercive power over research in the social sciences (Reichert, 2015), we should ask ourselves how humanists can study citizens’ cultural habits, in constant dialogue with libraries and museums and using methods to anonymise data. In my view, we should aim to ensure that companies like Spotify and Amazon do notknow more about a particular society – about our tastes, interests and moods – than its own members do.BIG DATA IN THE DIGITAL HUMANITIES · ANTONIO ROJASSmart culture. Analysis of digital trends