implicates digital rights, especially when interpreting those rights in local jurisdictions. In addition, the ad hoc nature of establishing and interpreting digital rights through the law also means that our understanding of legal frameworks for digital rights at any level is far from comprehensive. With all this in mind, we drafted a working definition of digital rights that attempts to capture their interdisciplinary, multidimensional, evolving nature. We wanted to highlight that these rights cannot be traced to a single authority or source, but rather are a product of distributed work, like the charters themselves. Further, we wanted to recognise that the spaces in which digital rights exist, like human rights, are unbounded. Thus, we adopted the phrase “digitally networked” to encompass not just the browsable internet but other digital networks. This is becoming even more important with the growing recognition that even people who are not connected online are increasingly affected by what happens in the digital sphere.33 Finally, we wanted to acknowledge that rights can be situated not just in content and interaction but also in other protocols, such as algorithms, on these networks or at their nodes, which come in the form of objects (devices) and in the form of expressions of our identities, whether individuals or groups, hidden, imagined, or in plain sight. Below is the definition we drafted. As a cornerstone of the refined methodology, it was meant to establish a reference point by which one can judge whether a law affects digital rights. It is a work in progress. Working definition: “Digital rights” describe human rights – established by the Universal Declaration of Human Rights, UN resolutions, international conventions, regional charters, domestic law, and human rights case law – as they are invoked in digitally networked spaces. Those spaces may be physically constructed, as in the creation of infrastructure, protocols and devices. Or they may be virtually constructed, as in the creation of online identities and communities and other forms of expression, as well as the agency exercised over that expression, for example, management of personally identifiable data, pseudonymity, anonymity and encryption. Such spaces include but are not necessarily limited to the internet and mobile networks and related devices and practices. 33 Tufecki writes “‘digitally networked movements’ or ‘networked movements,’ does not mean ‘online-only’ or even ‘online-primarily.’ Rather, it’s a recognition that the whole public sphere, as well as the whole way movements operate, has been reconfigured by digital technologies, and that this reconfiguration holds true whether one is analyzing an online, offline, or combined instantiation of the public sphere or social movement action.” Our working definition of digital rights served as a touchstone as we developed the rest of the methodology. In particular, it helped us devise a strategy for locating relevant legislation and then categorising that law. Step 3: Establishing criteria and a research path to identify relevant legislation Just as digital technologies have been integrated into every aspect of life, we can expect them to appear in multiple and increasingly diverse areas of law, from constitutions that make internet access a right, to health care laws that aim to protect patients’ data privacy, to anti-terrorism laws that restrict speech glorifying violent extremism on online platforms. Radar Legislativo,34 a legal data initiative from Brazil that tracks draft laws, recently counted 303 bills that affect the internet under review by that country’s National Congress.35 So even with the working definition in hand, we still needed to set criteria to help researchers narrow the field of inquiry and also give them a reasonable degree of certainty that the laws they found were in fact the laws they were looking for. To do this, we employed two complementary strategies: first, we looked at how Greenleaf identified data privacy laws, and second, we tried to locate the most likely areas in a legal framework where a researcher would find laws related to digital rights. In Greenleaf’s model, a researcher could identify a data privacy law in one of three complementary ways. First, they could look for laws that address data privacy principles, as defined by a “‘strong consensus’ that has emerged as to what are a set of twelve ‘fair information principles’.”36 Even a law with provisions that address only some of the principles could qualify the law as a data privacy law. To apply this to the problem of identifying a digital rights law would mean identifying the kinds of laws that routinely affect digital rights, or that are designed explicitly to establish norms for digitally networked spaces. While we know of no “strong consensus” about what laws might comprise a list of digital rights-related laws, we can deduce from the laws we and others have collected that it would likely include data privacy laws, right to information laws, etransactions laws, anti-cybercrime laws, and broad internet laws like Brazil’s Marco Civil da Internet (Civil Rights Framework for the Internet).37 34 https://www.radarlegislativo.org 35 Conversation with Kimberly Anastácio, Coding Rights, 18 October 2017. 36 Greenleaf, G. (2014). Op. cit. 37 www.cgi.br/pagina/marco-civil-law-of-the-internet-in-brazil/180 12 / Unshackling Expression