Many people consider accessing Internet services is a basic human right (as shown in an Internet Society survey in 2012). A report from the United Nations Human Rights Council states that “facilitating access to the Internet for all individuals, with as little restriction to online content as possible, should be a priority for all States”. In Horizon 2020, one of the seven priorities of the Digital Agenda for Europe is fast and ultra-fast Internet access. Besides creating the conditions to improve the coverage and quality of both fixed and mobile broadband Internet access, the Digital Agenda lists actions for member states to develop national broadband plans and regulatory measures in order to strengthen non-discriminatory Internet access (or network neutrality), to promote competition and to increase innovation.
Four years ago, Inria submitted a EQUIPEX Project called METROSCOPE. This project was not funded and METROSCOPE evolved into an open informal group of academic, industrial and social researchers and organizations interested in studying the Internet. Our objective is that METROSCOPE becomes, in the future, a web portal federating the main Internet measurement projects into a multi-dimensional Internet measurement observatory (from core to border metrics). The BETTERNET project is focused on Internet service access from User Experience (border metrics) and will use the METROSCOPE web portal to disseminate its research results. So we hope, thanks to the IPL, to attract external national and international partners interested in sharing and integrating their measurement tools and/or data. The BETTERNET will give the means to study net neutrality from the citizen’s perspective and will arm the Internet community to better ensure and the respect for privacy
Net neutrality is indeed one of the most frequently debated subjects concerning the regulation of Internet infrastructure and usages. The definition of this concept is that traffic should not be discriminated between different applications, contents or terminal equipments. One of the key issues related to net neutrality is the creation of a fair value chain across the Internet ecosystem. On the one hand, OTT players, such as YouTube4 and Netflix5, base their business models on advertisements and e-commerce without incurring costs related to the physical infrastructure of the Internet. On the other hand, network operators must invest massively in new capacities that are required to support the exponential traffic growth due to the ever growing demand for bandwidth aggressive contents, such as video, games or virtual reality, without any corresponding value creation as they use a “flat rate” model. This model is not sustainable and
another pricing model would rather be to charge for QoE. Indeed, some customers may be willing to pay for an improved QoE when using specific bandwidth consuming applications. They may also be willing to know the part of incoming and outgoing unwanted traffic (advertising, private data. . . ).
In this case, it becomes crucial to be able to verify the QoE that is effectively experienced by end-users, and in case of service degradation to locate the source of the problem (e.g., set-top-box, access network, interconnections between providers, or CSP). Unfortunately, it is challenging to measure QoE directly, we must then measure QoS (which is possible to monitor automatically with network level or application level metrics) and then infer the user’s experience with the application—i.e., the QoE. This research project therefore intends to provide a way to evaluate techniques for monitoring QoS and mapping QoS to QoE. It will also give a scientific framework which could help Internet users or national regulators to evaluate the way ISP or CSP handle traffic and what kind of discrimination they may apply. Our approach is open and transparent: through the concept of open data, collected data will go back to their owner.
