The Next Attack on Net Neutrality

One of the depressing things about net neutrality is that it is a battle that must be won again and again. It's becoming increasingly clear that another effort will be made by telecoms companies to destroy net neutrality at the big World...

Share

One of the depressing things about net neutrality is that it is a battle that must be won again and again. It's becoming increasingly clear that another effort will be made by telecoms companies to destroy net neutrality at the big World Conference on International Telecommunications (WCIT). Here's how it describes itself:

This landmark conference will review the current International Telecommunications Regulations (ITRs), which serve as the binding global treaty outlining the principles which govern the way international voice, data and video traffic is handled, and which lay the foundation for ongoing innovation and market growth. The ITRs were last negotiated in Melbourne, Australia in 1988, and there is broad consensus that the text now needs to be updated to reflect the dramatically different information and communication technology (ICT) landscape of the 21st century.

But the Internet Society sees things rather differently:

The Internet Society believes that decisions made by governments at WCIT could redefine the international regulatory environment for the Internet and telecoms in the 21st century and beyond – impacting how people around the world are able to use the Internet.

Modifications to the ITRs could result in changes to the Internet's architecture, operations, content and security. These are the kinds of issues that the WCIT may consider:

Peering arrangements and the impact on costs of international Internet traffic, which may change the way users pay for Internet services today;

"New technologies" regulation, which may open the way for censorship through technologies like DNS filtering that fragment the global Internet;

Data privacy, including access by the state to what is considered private data today and owned by citizens or organizations;

Cybersecurity, to give states more control over content and access to networks;

Internet addresses, which may lead to changing the global address registry and how users access websites today;

Mandated application of ITU-T recommendations, which may lead to slowing innovations, the diversion of technical resources from organizations, and fundamentally alter the open multi-stakeholder process responsible for developing the Internet today;

Misuse, fraud and spam

Clearly, the stakes are extremely high. The EU could play an important role in preventing some of the worst outcomes, but as this excellent post from La Quadrature du Net explains, on the key area of net neutrality, the European Union is extremely feeble, making only token statements of support for the idea.

One reason for that is because it is being lobbied hard by European telecoms companies, which see the WCIT meeting as an opportunity to take control of the Internet and refashion it along traditional telecoms lines. One body trying to make that happen is ETNO, The European Telecommunications Network Operators' Association. It has recently published a paper entitled "Contribution to WCIT: ITRs Proposal to Address New Internet Ecosystem" [.pdf].

The title alone says it all: "ITRs Proposal to Address New Internet Ecosystem ". As you might have noticed, this "new" Internet ecosystem has been around for a few decades; it's more that the telecoms companies have finally realised that not only is it not going away, it's going to completely replace their traditional business and its fat margins. This means they need to find a way to take it over and turn it into something they can make plenty of money from as they have in the past with the telephone system.

Here are more of ETNO's thoughts on the nature of this new-fangled Internet thing:

The telecommunications market and the telecoms industry as a whole is undergoing a fundamental shift. Catalysed by the availability of higher bandwidth connectivity, new applications and services are being enabled that go far beyond the traditional services of voice calling. In both the consumer and enterprise segments, services such as Voice over IP (VoIP), social networking, instant messaging and the rise of ‘apps' have changed the way customers use their mobile and fixed connections. This development is significant and telecoms operators need to adapt and rebalance their tariff structure between voice and data services.

From a technical point of view, networks are evolving towards an ‘all€?IP' platform. There is a progressive shift from the current model of separate ‘circuit switched' and ‘packet switched' networks (for voice and data services respectively), to a single future€?proof ‘all€?IP' network, supporting all services that the operator offers. At the same time, the increasing availability of fibre to the end€?user is stimulating the demand for new services and bandwidth consumption.

The first paragraph is basically an admission that telecoms companies won't be able to charge outrageous prices for voice calls, because they'll just be IP packets that cost an ever-diminishing amount to send. This means that telecoms companies must "adapt and rebalance their tariff structure" - in other words, find ways to charge outrageous prices for data instead. The rest of the document explains in detail how they aim to do that.

Basically, they want to kill net neutrality – which is what drives prices of individual IP packets ever-closer to zero because of the equivalent of Moore's Law for comms – so that they can start charging for complicated, non-transparent bundles of new services.

First of all, the telecoms companies have to invent a problem that requires the death of net neutrality as the solution. That problem, it seems, is that successful Internet companies like Google and Facebook are just too successful, and are sending too much data over the Internet. Despite the fact that they pay telecoms companies for this privilege, at prices set by the telecoms companies, it seems the latter can't run their businesses efficiently enough to make enough money in this sector to invest in new infrastructure. So the obvious solution is start charging companies like Google and Facebook extra to compensate the telecoms' incompetence. Or, as they put it:

Today there is a huge disproportion amongst revenues and a clear shift of value towards players (Over the Top players €? OTT) who are not contributing to network investment. Traffic and revenue flows need to be realigned in order to assure the economic viability of infrastructure investment and the sustainability of the whole ecosystem. The revision of the ITRs offers a unique opportunity to propose high€?level principles for IP interconnection.

You've got to love that term "Over the Top players" applied to Google et al. - no attempt to skew the discussion there, no sirree....

So what's the solution? Well, you guessed it: net neutrality must die – except we won't exactly phrase it like that...

ETNO wants to avoid decisions that would prevent new business models from emerging or that would hamper differentiated offers, hence limiting consumer choice. The risk of undesirable economic and technical regulation of operator rates, terms and conditions will be much higher if the development of the Internet continues to be jeopardized by the lack of sustainability and/or by the lack of end€?customer satisfaction.

That is, we must allow differentiated offers – aka no net neutrality – because of that "lack of sustainability" caused by telecoms operators' failure to adapt their own networks to the digital world fast and deeply enough, and, of course "end-customer satisfaction".

Except, of course, that preserving customer choice is why net neutrality must be not just kept but required by law. Net neutrality ensures that all IP packets are treated equally; that means that any new Net-based services are treated just like existing services. Which means that incumbents aren't able to squeeze out newcomers.

But that's precisely what they can – and surely will – do if net neutrality is destroyed. It will allow large search companies to pay for their services to be delivered rather more speedily than new, possibly open source ones; it would allow large publishers to have their content prioritised over that produced by bloggers. In other words, the absence of net neutrality is all about making the powerful more powerful, and the rich yet richer. It is not about creating a level playing field for innovation, as the Internet has been until now. The death of net neutrality would be the death of that unbridled creativity.

But ETNO thinks that's OK, because what users would get in return for losing all those innovative new services would be – drumroll – Quality of Service (QoS) differentiation:

QoS mechanisms allow differentiating services and types of traffic as well as specific content according to the economic value they represent to consumers and producers (OTT players or CPs). As an example, we can consider the delivery of video on demand: video on demand has a value for the CP who is selling the content and for the consumer. The quality of the data streaming is the essential factor in ensuring a good viewing experience. Content Providers have a strong interest in end€?to€?end quality in order to meet the demands and expectations of their customers. Hence, they have already started negotiating with telecoms operators on the modalities of quality assurance.

"Best effort" is generally sufficient for search queries.

There, in a nutshell we have it. You may think that those blogs you read are valuable, but ETNO knows that online films are much more valuable – because telecoms companies will be able to charge content providers a premium for delivering them. What that clearly signals is anything that doesn't have a Disney or News International behind it will be consigned to the Internet slow lane – delicately called "best effort" here.

But that's the problem: if the "best effort" turns out to be dog-slow, well, that's just tough: the telecoms company concerned will point out that it tried its "best", but unfortunately all its shiny new connectivity was taken up with video on demand, and there simply wasn't any speedy bandwidth left for the low-life Web sites that couldn't pay for access to the fast lane like the big boys.

And contrary to what the telecoms companies would have you believe – that net neutrality must be sacrificed on the altar of network upgrades – without net neutrality, there is zero incentive to improve the "best efforts" connectivity – the open Internet. Indeed, there are good reasons to make it as awful as possible, to drive people to pay extra for "QoS" services. So if politicians really believe their rhetoric about the importance of the Internet, and need for fast broadband connectivity, then they must protect net neutrality: that way, telecoms companies would be forced to upgrade the entire network, not just bits of it.

Of course, ETNO has compelling reasons why net neutrality must die – because if it doesn't, then you might:

Today's best effort only Internet provides no means to guarantee a specific level of traffic delivery. In a most benevolent interpretation this means that "inelastic applications in fact work tolerably well over a best€?effort network", as the Internet Society has put it in their position paper . This broad statement is not only assuming that bandwidth may be increased indefinitely and ahead of actual demand but also neglecting the fact that certain applications, such as e€?health cardiac monitoring services for example, may not depend on "tolerably well" transport but need to be quality€?assured. There are inelastic applications that may never be realized in a best effort only Internet. Thus, extending the technical options is a necessary prerequisite for further innovation.

Yes, it's the tired old "what about the cardiac monitoring services?" line. But again, this is not an argument for destroying net neutrality, but for letting competition between telecoms flourish to produce improved, more reliable connectivity for the whole, open Internet – something that won't happen if "premium" services allow them to cherry pick only the most profitable sectors for their offerings.

And as I've written before, if users want to prioritise their video on demand or gaming over those boring old Web searches, then let them – but the control must rest at the user's end, perhaps through a simple dashboard application that lets them set how much of their undifferentiated bandwidth is allocated to what application and for how long. That way, there's no need for intrusive Deep Packet Inspection that QoS necessarily implies. There's also no way for powerful companies to tilt the playing field to the disadvantage of newcomers, and so innovation would be able to flourish unhampered.

At the end of the ETNO paper, the mask drops, and it becomes apparent that what is wanted is to turn the Internet into the old telephone system. That's shown by the fact that the paper goes out of its way to deny it:

By proposing a charging principle that is well tested and has delivered efficient results ETNO is not trying to bring back the circuit switched architecture of the PSTN. The main aim is to promote , where appropriate, a charging principle which re€?aligns costs and revenues by assuring that the agents which take the economic decisions – in this case the routing decision – are accountable for the costs they incur.

Again, this is a clear admission that those who could afford it would be able to get their IP packets sent faster than those who couldn't – regardless of what the end user wants. But the Internet has thrived because it has done the opposite: the user gets to choose how fast they want their connections to be, and the Internet companies merely meet that demand. Crucially, it has been a user-driven revolution.

At the end, the document says that net neutrality must be subordinated to "commercial negotiations":

While all Internet service providers strive to achieve "universal connectivity" - which means any device may communicate with any public IP address – the commercial reality is, that this is hardly ever fully achievable. ETNO is not asking for increased governmental intervention but stresses the priority of commercial negotiations.

In other words, rather than the technological egalitarianism that is inherent in the very nature of TCP/IP, ETNO wants money to be the ultimate arbiter of what gets sent where, to whom and how fast. Like the many powerful companies frightened by the decentralised power of this medium and its proactive users, ETNO wants the Net neutered, and its dangerously independent community of independent users turned into a bunch of passive consumers choosing from a limited menu of "quality" services.

Follow me @glynmoody on Twitter or identi.ca, and on Google+

Find your next job with computerworld UK jobs