On 29 May 2019, EDRi was invited to participate in a workshop of the Body of European Regulators for Electronic Communications (BEREC) on the planned update of its Net Neutrality Guidelines. Thomas Lohninger from Austrian EDRi member Epicenter.works and Jesper Lund from Danish EDRi member IT-Pol represented our network. Lund provided the following input to the regulators on regulatory action by the National Regulatory Authorities (NRAs).
Epicenter.works published a report in January 2019 which, among other things, surveys regulatory action based on the annual net neutrality reports by the NRAs. Port blocking is a severe form of traffic management since entire services, such as hosting of email or web servers by the end-user, are suppressed. This may be justified in certain situations, but requires a rigorous assessment under Article 3(3) third subparagraph, point b (preserve the integrity of the network) of Europe’s Net Neutrality Regulation (2015/2120).
Port blocking is generally quite easy to detect with network measurement tools. This is also noted in section 4.1.1 of BEREC’s Net Neutrality Regulatory Assessment Methodology (BoR (17) 178). Other forms of discriminatory traffic management are harder to detect. Based on this, it seems a reasonable conjecture to take NRA enforcement action on port blocking as indicative of the rigorousness of wider enforcement practices regarding traffic management. Unfortunately, detailed information on port blocking cases is not contained in most NRAs’ net neutrality reports.
Since the publication of the Net Neutrality Guidelines in August 2016, BEREC has launched a project to create an EU-wide network measurement tool, expected in late 2019. The measurement tool is based on the core principles of open methodology, open data, and open source. This means that the tool can be deployed on many devices, used by many end-users, and that the data generated through “crowdsourcing” by end-users (subscribers of internet access services, IAS) can be analysed by NRAs and other interested parties. In the opinion of EDRi, effective use of the forthcoming measurement tool, with crowdsourced measurement by end-users, will be a milestone in supervision and enforcement actions for traffic management practices.
Among other things, the measurement tool can be used for detection of unreasonable traffic management practices, establishing the real performance and Quality of Service (QoS) parameters of an IAS, assessing whether IAS are offered at quality levels that reflect advances in technology, and assessing whether the provision of specialised services risks deteriorating the available or general quality of IAS for end-users.
All of these tasks are specific obligations for NRAs under the Open Internet Regulation. As EDRi has highlighted before, the crowdsourcing aspect of the deployment of the measurement tool is very important as single measurements can contain a large element of noise, for example because of characteristic of the specific testing environment. In the aggregate, the noisy element can be expected to “wash out”, leaving the effect of the IAS traffic management practices or other network design choices by IAS providers.
When a measurement tool developed by BEREC is freely available to NRAs, the Guidelines on Article 5 of the Regulation should be updated to contain specific requirements and recommendations for the use of network measurement tools in the NRA supervision tasks. NRAs should, of course, be free to choose between their own measurement tools and methodology and the one offered by BEREC to all NRAs.
The Regulation does not per se require NRAs to establish or certify a monitoring mechanism. Needless to say, the Guidelines cannot change that. Therefore, most provisions in the Guidelines related to network measurement tools will have to be recommendations for NRAs.
However, the Regulation specifically requires NRAs to closely monitor and ensure compliance with Article 3 and 4 of the Regulation. While NRAs should be free to choose their own regulatory strategies, allowing these strategies to be adapted to the local “market” conditions and need for enforcement action, some proactive element is required on behalf of NRAs. Simply responding to end-user complaints cannot be sufficient to satisfy the obligation under Article 5.
In the opinion of EDRi, it will be very difficult for NRAs to fulfil their monitoring obligations under Article 5 without some form of quantitative measurement from the IAS network. The last sentence of recital 17 of the Regulation oncretely requires network measurements of latency, jitter and packet loss by NRAs to assess the impact of specialised services.
BEREC’s Guidelines with recommendations on the use of crowdsourced network measurements will have two positive implications for the net neutrality landscape in Europe. For the NRAs that follow the recommendations, and actively use the BEREC measurement tool, we will have quantitative monitoring of the compliance with articles 3 and 4 that is harmonised and comparable across EU Member States. This will, in itself, be hugely beneficial, and contribute to a consistent application of the net Neutrality Regulation.
In Member States where the NRA decides not to use the BEREC measurement tool (or its own), the recommendations in the Net Neutrality Guidelines could potentially facilitate shadow monitoring reports by civil society or consumer organisations. Of course, this can also be done without recommendations in the BEREC Guidelines or even with alternative measurement tools (than the one developed by BEREC), but adhering to the BEREC recommendations would create results that can be more easily compared with for example NRA net neutrality reports in Member States where the BEREC measurement tools is actively used.
EDRi will be pleased to contribute draft amendments to the Guidelines in order to formally incorporate a network measurement tool and crowdsourced measurements in the IAS network by end-users.
BEREC Workshop on the update of its Net Neutrality Guidelines
Europe’s Net Neutrality Regulation (2015/2120)
BEREC Net Neutrality Regulatory Assessment Methodology
BEREC Guidelines on the Implementation by National Regulators of European Net Neutrality Rules
Two years of net neutrality in Europe – 31 NGOs urge to guarantee non-discriminatory treatment of communications (30.04.2019)
NGOs and academics warn against Deep Packet Inspection (15.05.2019)
(Contribution by Jesper Lund, EDRi member IT-Pol, Denmark)