Introduction

Net neutrality (also known as the open internet) is one of the requirements for a functioning global network. It implies that all electronic traffic passing through the network should be treated equally, with no discrimination (as far as possible – though technically difficult to achieve, if there are different tariff plans), regardless of the type of content being sent, or application or service by which the message was sent, or devices from which it was sent, or the address of the sender or recipient. Briefly, it consists in the fact that network users may simultaneously use different internet services, while providers cannot discriminate between traffic to or from any portal.

The importance of neutrality is best illustrated by an example. Imagine that the postal service decides to deliver only some letters or that a telecommunications network operator connects only selected telephone calls. It would be dangerous, too, to operate a selective delivery of water, gas or electricity to residents. The same is true with internet access. Suppliers could make one page open more slowly than another, or not launch at all.

How may traffic parity be reduced?

When discussing the principle of neutrality,  one should start with the role of internet service providers (ISPs). They typically use three methods for limiting the flow of data on their links: (i) throttling, (ii) blocking, and (iii) privileging paid traffic (in order of strongest to mildest).

Blocking may take an extreme form and involve refusing a connection to the web. But this does not happen very often these days. Nonetheless, it is possible that an internet service provider may block access to particular sites.

Traffic throttling is used to avoid accusations of blocking. The internet service provider enables technical access to an internet service, but curbs the speed of the transfer so much that effective use is not possible. Throttling may be imposed because of the source (e.g. content from a particular supplier), target (e.g. to a particular consumer) or content (e.g. pornographic videos). Therefore, in terms of effect, throttling is synonymous with blocking.

Blocking and throttling are, in general, both prohibited. However, this is a relative ban that is limited by the rule of sound net management - one of the powers of an ISP, as a manager of infrastructure. However, this exception does not, though, apply to a third type of activity that is purely commercial and technical and maintains the net’s efficiency.

Paid privileging describes the activities of a broadband access provider consisting of direct or indirect acceleration of traffic in one service with respect to another. ISPs can use tools such as: traffic modelling, prioritisation, resource reservation, and other types of traffic management in return for remuneration or to benefit a party associated with the ISP.

Net neutrality principle - who decides the form of the internet?

The principle of net neutrality is intended to counter such ISP practices. It was only laid out in legal acts in the 21st century, while the idea itself of the neutrality of communications service providers was born in the United States as early as in 1860. The Pacific Telegraph Act, concerning telegraph lines and communications, included for the first time a provision that addressed the problem: “messages received from any person, firm or corporation or from any telegraph line connecting that line at any end should be sent impartially, in the order in which they were received [...].”

As for Europe, neutrality was usually guaranteed by telegraph networks’ rules of operation. This rule was enshrined in the first international telegraph convention and in telegraph regulations. The postulate of impartiality was more of a technical necessity, rather than ideological, because different networks were unable to connect with each other.

The first country in the world that committed its legal system to respect the principle of net neutrality was Chile (see here and here). Before its introduction at the EU level, this rule was explicitly regulated in only a few national legal systems, e.g. in the Netherlands, Slovenia and the United States.

The debate on net neutrality has been rolling in the United States for more than twenty years. The first act that included net neutrality provisions was the Internet Freedom and Non-discrimination Act of 2006. It forbids modifications to transmitted content and limiting transmissions, except for viruses or spam. There is an ongoing dispute over the regulation of this issue. After a long discussion, the Federal Communications Commission (FCC), a federal government agency, adopted rules for promoting and protecting the open internet on 26 February 2015. The Commission decided to recognize ISPs as common carriers, which vindicated the adoption of net neutrality. Their principle objective is forbidding ISPs from blocking, throttling or charging for preferential traffic. Suppliers also may not offer “fast lanes”, namely a service offering selected network users better access to specific content. Those rules entered into force on 12 June 2015. However, due to the appointment of a new FCC President, Ajit Varadaraj Pai, consideration is being given to the early repeal of rules on net neutrality in the United States.

The above rules were implemented under two acts: the amended Title II of the Communications Act of 1934 and Section 706 of the Telecommunications Act of 1996. This first document defines the common carrier concept and in Section 202 prohibits any discrimination of internet users[1]. Despite the dissenting views of operators and telecommunications corporations following the entry into force of the open internet rules, they were upheld in force by the Federal Court of Appeals for the District of Columbia in a judgment dated 14 June 2016

In the European Union, initial issues with telecommunications services and operators’ transmissions of data were regulated by the five Directives of 2002. Over time, specific rules were devised on net neutrality. The Body of European Regulators of Electronic Communications (BEREC) was established in 2009 under Regulation 1211/2009, which brings together electronic communications market regulators from the Member States. The President of the Polish Office of Electronic Communications (UKE) is one of the members of BEREC.

Since recently, the issue of net neutrality is regulated by EU Regulation 2015/2120. Article 3 par. 3 of the Regulation establishes the principle of net neutrality within the European Union: “Providers of internet access services shall treat all traffic equally, when providing internet access services, without discrimination, restriction or interference, and irrespective of the sender and receiver, the content accessed or distributed, the applications or services used or provided, or the terminal equipment used”. However, because the wording of the Regulation’s provisions is not clear, and the operators and service providers that relied only on them could have infringed net neutrality, on 30 August 2016, BEREC issued guidelines for the interpretation of Regulation 2015/2120.

Why are guidelines so important? In EU Member States, the national supervision authorities are tasked with implementing net neutrality, and BEREC brings them together. Therefore, what issues did BEREC refer to in its guidelines and what caused the largest contentions? First and foremost, Regulation 2015/2120 did not show how neutrality should be interpreted in the context of three issues: traffic management, specialist services and zero-rating. All three contravene net neutrality but may be justified. Traffic management by ISPs involves privileged treatment of selected users. Specialist services is a concept promoted by Chancellor Angela Merkel. With a division into the free internet and specialist services, operators will better serve certain key sectors such as telemedicine and autonomous vehicles. As for zero-rating, this means that selected services are offered free, as they do not count towards purchased data transfers.

So how did BEREC deal with the above issues? The guidelines show clearly that infringements of net neutrality are prohibited, but with certain exceptions. Zero-rating is permissible, in certain situations, such as when applying a zero rate to an entire category of applications. Traffic management is acceptable only if it is transparent, non-discriminatory and proportionate. In accordance with the guidelines, equivalent categories of traffic should be treated equally: suppliers may vary the speed of traffic but only between objectively different traffic categories. On the other hand, specialist services may be provided for reasons of optimisation, to meet the requirements for a specific quality of service level. The guidelines show that the EU will be striving for stricter regulation of neutrality and departures from it must have a solid basis in the need and proportionality of the action.

What does the internet of things bring that is new?

The status of the law laid down in the United States and the European Union in recent years may have to change due to further technological advances. This is due to the spread of the internet of things, which will very likely force certain changes in the legal regulation of net neutrality.

The internet of things involves the spread of embedded systems, namely technologies that closely tie hardware with software. This means bringing the digital world into tangible reality, as a result of which there will be a blurring of the distinction between virtual and real worlds. This implies the emergence of computer programs that will combine with everyday consumer devices, which will be equipped with fixed internet connections and will often have their own version six IP addresses (IPv6). Ipv6 requires the creation of 128-bit addresses (current IPv4 addresses are 32-bit), enabling the individual addressing of billions of new devices.

Smart devices, which make up the internet of things, may be divided into two groups. The first includes all those that are used in industry. Example applications include those for tracking movements of goods in logistics and automatic fault detection. Devices in this group will not always be part of the internet of things, because the movements of data may be restricted to a closed circuit in the factory or within the device (e.g. an electricity generator or an airplane). Only once the data is transferred outside the closed circuit, via the internet, is it possible to refer to it as the internet of things.

In contrast, the second group should include consumer devices, such as smart watches, clothes with sensors, medical diagnostic devices and smart domestic appliances such as refrigerators and thermostats. This group should also include autonomous cars, which are being currently tested on some countries’ roads.

The proliferation of the internet of things is being largely supported by developments and technologies already present on the market, such as big data, smart grids and autonomous machines.

Data collected and uploaded by the internet of things

Smart devices will be collecting enormous amounts of data. Most of those devices will be connected permanently to the internet, or will connect regularly from time to time, or once any event has been detected. This will enable automatic exchanges of data between devices (machine-to-machine) and the making of decisions about the use of data independently collected or received from outside – all, to a certain extent, beyond human control. That is the reason for the “internet of things” term, which multiplies the scope of uses of the “internet of people”. Other identical notions are the internet of objects, the internet of everything, and pervasive or ubiquitous computing.

Obtained data will sometimes appear in raw form and will be sent in such form, if the internet of things devices are not very sophisticated. This is due to the current state of miniaturisation technology and common and cheap internet access. It is therefore worthwhile for entrepreneurs to build simple diagnostic devices that will then transmit data over the net. The uploading of data over the internet from the consumer to the manufacturer or between machines constitutes the point of contact between the internet of things and the principle of net neutrality.

Who loses, who gains?

It is possible to distinguish several groups that are currently interested in the problem of net neutrality in view of the approaching internet of things. Those groups are: (i) users, (ii) device manufacturers, (iii) social networking sites, (iv) application developers using application programming interfaces (APIs), (v) content providers, and (vi) internet services providers (ISPs). Which of these will gain and which will lose on a possible abolition of net neutrality?

Primarily, it is clear that the interests of users and ISPs are at loggerheads. For users it would be advantageous to maintain net neutrality, as this rule enables web surfing, which really means unrestricted and immediate switching between services from different content providers. On the other hand, ISPs would benefit more from privileged traffic, because currently they earn little in comparison with content providers.

The group of net neutrality beneficiaries may include, alongside users, also device manufacturers and developers of external applications. These are developers, which are independent of content providers and ISPs, that identify new uses for the internet of things. They propose new services such as the increasingly popular lifelong or quantified self monitoring devices that are used for medical or fitness-building purposes. It is understandable that in creating a new resource, they will want an open net for their potential users.

As for the group of net neutrality opponents, alongside ISPs, this may include also social networking sites and content providers. With regard to the latter, this is, of course, about service providers with content that puts a strain on data traffic, namely streaming services.

It would seem that the forces have lined up three against three. Nevertheless, on one side, there are the users, independent app developers and device manufacturers, and on the other social media providers, content providers, and ISPs. It is clear that the forces are matched unevenly. This asymmetry explains why the US federal government and EU institutions have decided to intervene. Is this legitimate?

Looming problems

There were no doubts about net neutrality in the “people's internet”. Could it be maintained in the internet of things? It seems very unlikely.

The increase in the popularity of internet of things devices will make internet traffic incomparably greater than today. At the same time, machine-to-machine autonomous communication will increase the need for the reliability of certain services, meaning permanent and fault-free internet access. Already there is the example of an autonomous car causing a fatal crash after a broken connection. Similarly, data that could be essential for people’s safety could be transmitted from smart electricity meters thus enabling rational management of the energy grid during a crisis. Therefore, in an internet of things world, it is not going to be easy to reject Angela Merkel’s postulate for special services to be prioritised.

Apart from special services, devices that provide trivial services will also transfer data: for example, gadgets that closely track progress in training; likewise with many smart household appliances. The data generated by those sources is not essential for public order or safety. Therefore, it is possible to imagine traffic choking due to such devices. This, however, would involve having to abolish the principle of net neutrality.

In view of this need, solutions may have to be sought elsewhere - in areas of law with a longer tradition than the principle of network neutrality, e.g. in competition law and in regulated markets law. The institutions of the former could protect competitors from abuse of dominant position or even suggest dividing the holder of such a position. This latter approach has been used in the past in the very home of the free market economy and of competition law. The most famous example is the division of Standard Oil Co. Inc., the company created by John D. Rockefeller, into many smaller companies, which are still operating today. On the other hand, legal tools that have originated in regulated markets (such as telecommunications, power, post and rail) suggest the option of imposing public obligations on enterprises. Therefore, maybe one should, actually, resort to those tried-and-tested tools for the new entities born from the internet boom of the century’s first decade?

Aleksandra Lisicka, Wojciech Rzepiński