Net neutrality has been in the news a lot lately, but there’s much less discussion of its relationship to and impact on cloud computing (or, for that matter, the Internet of Things).
Recent developments in net neutrality
On February 26, 2015, the FCC in the United States ruled in favor of “net neutrality” by reclassifying broadband access as a telecommunications service and applying Title II (common carrier) of the Communications Act of 1934 to Internet service providers (ISPs). FCC decision FCC-15-24A1, a massive 400 page document, was released publicly on March 12th. The lawsuits, which were fully anticipated, are already starting to be filed.
According to the Public Knowledge website states (I would have thought this also applies to organizations and applications, not just individuals!):
“Net neutrality is the principle that individuals should be free to access all content and applications equally, regardless of the source, without Internet service providers discriminating against specific online services or websites. In other words, it is the principle that the company that connects you to the internet does not get to control what you do on the internet.”
The entry in Wikipedia for net neutrality includes one of the longer reference lists I’ve ever come across on Wikipedia; this has to be a measure of the strategic importance of this topic.
IEEE Spectrum magazine has also had some interesting articles about net neutrality – including an assessment of the various metaphors that are often used.
In Canada, recent CRTC decisions are also favouring openness, flexibility and equality in communications service provision. The implications of net neutrality touch on a wide range of networking, broadcasting, and telecommunications topics. Internet-based systems would certainly benefit from worldwide coordination of net neutrality (after all, we have one global Internet, not one per country).
I suspect that many Internet pioneers are astounded (or at least highly amused) that their baby is now called a telecommunications service, and perhaps they are even surprised by all the twists and turns on the Internet highway.
What does net neutrality really mean?
As is often the case for IT terminology (e.g., what is cloud computing?), net neutrality has different meanings depending on your perspective, your scope of interest, and your marketing objectives. An IEEE Spectrum infographic describes four topics:
- Equal treatment for all bits;
- Equal treatment for all information providers;
- The industry’s view of net neutrality; and
- The technology behind net neutrality.
As noted by ZDNet, “Overall, net neutrality enforces three rules: no blocking, no throttling, and no paid priority traffic.”
Some basic questions come to mind:
- Is the Internet, and hence its access, truly a public resource (i.e., a utility) that must be available to everyone under the same terms and conditions? Is this true for both shared and dedicated network resources?
- Is Internet neutrality primarily a national matter, or is an internationally harmonized approach needed?
- What is the scope of the Internet (i.e., is it simply a transport service, or does it include the contents and applications both within the network and at the endpoints)?
- Should net neutrality apply only to the open, public Internet or to all TCP/IP-based networks? Can private networks share the same physical “wires” as the Internet?
- Do net neutrality rules affect the governance, management, performance and quality of the Internet? For example, how would breaches of net neutrality be detected and reported?
Net neutrality is based on two general technical principles that are inherent in today’s Internet standards:
- Best efforts delivery – the network attempts to deliver every packet to its destination equally, with no discrimination and provides no guarantee of quality or performance;
- End-to-end principle – in a general purpose network, application-specific functions should only be implemented at the endpoints of the network, not in intermediate nodes.
Net neutrality aims to ensure that the Internet is an open system that espouses and supports diversity, inclusiveness, non-discrimination, fair competition and innovation. Wouldn’t these goals also be desirable for public cloud computing?
Is net neutrality really that important?
Based on the number of people taking sides for or against net neutrality, there can be little doubt that it is important technically, commercially, and politically. Internet users should also be concerned about abuses that could occur without the safeguards that net neutrality proposes.
Some examples that illustrate how an ISP could violate net neutrality principles:
- Blocking – some users could be prevented from visiting specific websites or accessing specific services, such as those of a competitor to the ISP;
- Throttling – different treatment could be given to specific sites or services, such as slower speeds for Netflix;
- Re-direction – users could be automatically redirected from one website to a competing website;
- Cross-subsidization – users of a service could be offered free or discounted access to other services;
- Paid prioritization – companies might buy priority access to an ISP’s customers (e.g., Google or Facebook could (in theory) pay ISP’s to provide faster, more reliable access to their websites than to potential competitors).
From the ISP’s perspective, net neutrality places restrictions on potentially revenue-generating functionality. It may also impact how private networks might co-exist with shared public networks. Net neutrality enforcement can also be an important governance issue.
An Internet user should be able to connect to any other legal endpoint without interference from the service provider. This is analogous to ensuring every telephone can call every other telephone, anywhere, without restrictions on connectivity or quality. From the user’s perspective, net neutrality eliminates:
- Connection, admission and access discrimination;
- Paid prioritization or scheduling of access and/or transport; and
- Controls and limitations on applications and contents.
So, what about “cloud computing neutrality”?
Does neutrality apply equally to cloud computing or are they completely different issues? If net neutrality applies to (public) Internet services, then would cloud neutrality relate to public information processing and storage (i.e., SaaS) services?
The three rules of net neutrality also apply to cloud computing:
- No SERVICE blocking – SaaS providers should not arbitrarily restrict or block access to computing and storage services;
- No SERVICE throttling – SaaS providers should not favour one customer over another in areas such as capacity, elasticity, accessibility, resilience or responsiveness;
- No paid priority SERVICES – SaaS providers should not selectively offer (or provide) better services to selected customers at the expense of others.
For example, the following might hypothetically be possible:
- A SaaS provider could favour one search engine over another by preventing or slowing down search scanning;
- A SaaS provider could degrade response times for certain companies (such as a re-seller or broker) or users.
It would seem that the whole question of neutrality – the fair and open availability of public IT services – is more complicated it would be for other utilities such as water, roads or electricity. There is a need to look at net neutrality both holistically and technically as well as commercially and politically.
Will net neutrality be the driving force for major revisions to the traditional Internet standards?
Let me know what you think!