Net Neutrality for Dummies


by Mika Skarp

2015 has already proved a banner year for Net Neutrality with new regulations and legal decisions for both the USA and Europe in just the first 6 to 7 months. Pundits predict that other regions will quickly follow suit with their own language to deal with this very fluid issue.

As we look to the future of information technology and this important principle, it's good to try to refine our understanding of the subject. And it does bring some weighty questions; What is the future of free Internet? How can Application-Aware Network be neutral?  What does Net Neutrality actually mean?

To get there, we need to come together on what we might call "Internet First Principles".  First, is the fundamental need for a single, solitary multipurpose network called The Internet. There is consensus that this is essential to the idea of the internet as a truly inclusive global network and its ability to serve its defining role in pursuit of the public interest.  Aside from any other social, legal, political or ethical issues is the simple fact that the idea of introducing 'tiers' of the public internet would likely render them unaffordable. This will be looked at in greater detail later, but suffice to say that cost is one of the cornerstones of Net Neutrality.

A technical first principle of the internet, but one that only now beginning to come into full relief is the reality that different applications require different kinds of services from the network. The fact is that one size does not fit all when it comes to digital services. In FCC and EU regulation there are considerable tracts of text making reference to the unique demands of  IPTV, Video conferencing, Healthcare and the Internet of Things (IoT).

Following this logic, we can say that different digital services require different network parameters to deliver quality user experiences. This is the first key to understand Net neutrality; user demand is the primary driver for network parameters. Thus, users must be able to trust that the service delivers on the promised experience. Imagine for an instant (some of us may not have to), that the day in 2007 when Netflix launched its Video on Demand service that a standard high speed internet connection simple couldn't deliver. This is anathema to good business practice.

While there is a lot of internet traffic that can use so called "best effort" technology, there are also services that require strict Quality of Service (QoS) Service Level Agreements (SLAs) in different flavors. This is something that has been stated in both FCC and EU regulations; we should have different SLA classes in one network.

It is also logical that inside one service class, prioritization should not happen based on the content.

This is the second key to understand Net neutrality; all traffic inside one customer experience needs to be delivered equally. Here again, a good example is video streaming where, for example, Netflix and YouTube should be treated inside one SLA class equally.

How can we do this in networks?

To provide an excellent user experience for everyone we need to provide application-based network profiles. These profiles need to be such that a user can turn them on and off at will based on their need. In practice, this means that not only are computers 'cloudified' but networks are as well.

In mobile networks managing capacity is more challenging that in fixed networks, but with the emergence of 5G mobile networks this will become standardized and increasingly easier to manage.

To summarize, Net neutrality; one network, several application profiles, no prioritization based on content and services available for everyone on affordable price.