Net neutrality has in recent years become a prominent area of interest for BEREC – the Body of European Regulators for Electronic Communications – which is composed of high level representatives of the relevant national regulatory authorities (NRAs) in each EU Member State. In 2011 the European Commission asked it to undertake a fact-finding exercise on issues crucial to ensuring an open and neutral internet, and since then, BEREC has published various frameworks and guidelines on quality of service and transparency as well as findings on the traffic management practices and restrictions applied by ISPs.
This latest report follows its public consultation on internet access service (IAS) quality monitoring systems, and seeks to establish a basis for creating quality-monitoring that is capable of (i) enhancing transparency for end users of electronic communications services and (ii) assisting NRAs in assessing potential degradation of service and intervening with appropriate corrective measures. In this context, BEREC discusses issues relating to the metrics and methods of quality monitoring IAS, both as a whole and individually, and considers the costs benefits of controlled and less controlled measurement systems in different contexts. It recommends that:
- for the quality monitoring of IAS as a whole upload/download speed, delay and jitter, and packet loss ratio should (as a minimum) be measured at the IP layer, while for individual applications using IAS, assorted applications (i.e. content and applications) should be measured by NRAs to check whether there is degradation;
- for transparent information about IAS for individual end users, a software-based measurement agent downloaded to end user equipment can be sufficient, low cost and user-friendly, while for in-depth long-term regulatory supervision of IAS quality, a more controlled measurement system (e.g. with hardware probes) would be appropriate – NRAs will need to strike a balance between the two.
BEREC also goes on to consider the possibility of establishing a multi-NRA opt-in quality-monitoring system dedicated to regulatory purposes. This is no small feat given such a system would need to meet the current and future needs of NRAs as well as providing the basis for cross-border measurements. Although BEREC recommends an evolutionary multi-stage approach in principle, it recognises the potential practical challenges involved e.g. cost of cooperation, complexity of system, time constraints related to alignment among NRAs etc. It therefore proposes an initial feasibility study to investigate whether an opt-in approach involving the convergence of methods and sharing of infrastructure could be realised in practice, including NRAs with no existing systems.