Don’t outlaw successful business models on the Internet
This is the second article in the series on Net Neutrality: H.R. 3458 – A dangerous experiment in Internet regulation.
Representative Ed Markey has proposed some heavy handed Net Neutrality rules in H.R. 3458 that would outlaw several existing business models. The FCC may also be considering a subset of H.R. 3458 rules where the FCC would determine which managed services should be permitted on broadband and wireless networks, but it is unclear what the final rules will look like until they’re published.
Before we dive in to these newly proposed rules, some background information is useful to put everything in context. Google and other Net Neutrality proponents have long championed these heavy handed regulations on managed services on the Internet. Now that Congress and the FCC is considering whether to adopting them, it is imperative that we study the merits (or the lack thereof) of these proposals. Google has given us a great starting point by publishing their definition of Net Neutrality on what should and should not be permissible. Table 1 takes Google’s definition and highlights the contradictory positions.
Table 1: Google’s contradictions on Net Neutrality
|Providing managed IP services and proprietary content (like IPTV); and||Prioritizing data packet delivery based on the ownership or affiliation (the who) of the content, or the source or destination (the what) of the content; or|
|Employing certain upgrades, such as the use of local caching or private network backbone links;||Building a new “fast lane” online that consigns Internet content and applications to a relatively slow, bandwidth-starved portion of the broadband connection.|
|Charging consumers extra to receive higher speed or performance capacity broadband service.||Levying surcharges on content providers that are not their retail customers;|
Crippling managed services and private networks
H.R. 3458 attempts to regulate private networks and managed services. Here are three specific sections of the bill that attempt to accomplish this.
- to guard against discriminatory favoritism for, or degradation of, lawful content, applications, or services by network operators based upon their source, ownership, or destination on the Internet;
- ensure that private transmission capacity services do not undermine the purposes of this Act and do not diminish or degrade the level of Internet access service offered to the public by the same provider; and
- ensure that private transmission capacity services are not offered in an anticompetitive, unreasonable, discriminatory, or deceptive manner.
These provisions in H.R. 3458 seem to mirror Google’s concerns. Google is the largest online video distributor on the Internet which puts them at odds with traditional managed delivery television services. While Google says that providing managed IP services and proprietary content like Internet Protocol TV (IPTV) should be permissible, they want the means of providing those managed services to be outlawed. IPTV needs to be prioritized based on where it’s coming from so by prohibiting prioritized data packet delivery based on the ownership or affiliation, it leaves us with some brutally inefficient options. Google and other Net Neutrality proponents justify this by saying that managed services should only be permitted if it does not “diminish” or “poach” capacity from Internet capacity. While this might sound reasonable, it is easy to demonstrate why it is so foolish.
When we prohibit IPTV (or any other managed service like cellular phone service) from “diminishing” Internet capacity, the flip side of that coin is that Internet capacity is no longer allowed to use idle IPTV capacity even when no one is watching television or when fewer television sets are turned on. So if we went by this network architecture where resource is statically allocated to the Internet and IPTV, we end up with the grossly inefficient network architecture illustrated in figure 1 below where Internet capacity is stuck at 50% of the resources and no more.
Figure 1: Static resource allocation
In figure 2 below, we have the intelligent dynamic network resource allocation model which is presently used by every IPTV provider in the world. In this model, IPTV must always be given priority over Internet capacity to prevent Internet usage from disrupting IPTV, but the great thing about this system is that Internet capacity can now take 50% to 100% of the total capacity of the circuit. Yet Google and other Net Neutrality proponents see this as unfair discrimination. But if we’re going to permit IPTV at all, we either use static or dynamic bandwidth allocation and it’s clear that dynamic allocation is the superior solution. In fact, this concept of dynamic bandwidth sharing is one of main advantages of packet switching networks like the Internet over the old circuit switching model used by traditional communication networks like the telephone.
Figure 2: Dynamic resource allocation
When I presented this prioritized dynamic sharing model to Google’s Sr. Policy Director Richard Whitt at my network management presentation last year, Whitt deplored this prioritized IPTV model. But asked if he preferred the static sharing model shown in figure 1, Whitt liked that even less. Asked if Google would prefer that both models be disallowed, Whitt would not respond. Given the fact that Google’s business benefits greatly if consumers shift from traditional television services over managed networks to Internet based delivery, it’s a reasonable assumption that Google opposes any technology that competes with their own business. While there’s nothing wrong with any business looking to gain market share, regulators should not be in the business of favoring one business model over another by outlawing or crippling existing business models. If those new business models were so great, let them prove themselves in the market place by having consumers vote with their eyeballs and pocket books.
Note: These same principles of dynamic bandwidth sharing also applies to cellular phone service and wireless networks. Prioritizing cellular phone service over wireless data produces the same benefits in efficiency and it gives the data network more available capacity while ensuring quality on cellular calls.
Cable TV uses this type of static resource allocation because it has to support an existing architecture with legacy devices. Furthermore, the FCC requires cable companies to carry legacy analog television until the year 2012 even though analog TV broadcasts have ended. This need to support existing devices means that cable companies have to statically allocate 95% of their network capacity to cable TV service and only a few percent to Internet services. Since H.R 3458 applies to private capacity networks in addition to Internet based services to prevent them from “diminishing” the Internet and the FCC seems to be hinting at doing the same, it leads one to wonder if regulators will begin to demand taking capacity away from TV services in favor of the Internet even though more consumers still choose to pay for TV.
The myth that fast lanes should be illegal
Google and many other Net Neutrality supporters want to see a regulatory ban on “fast lanes” on the Internet. Yet this is completely contradictory to Google’s position that it is permissible to own “private network backbone links”. But what is an exclusive private network backbone that allows a company to bypass the congested Internet? How can this be defined as anything other than a “fast lane”? Put this in the context of Google now owning exclusive access to 7% of the Internet’s infrastructure and that it is among 30 companies that control 1/3 of the entire Internet. There’s nothing wrong with owning this much of the Internet when Google is paying dearly for all that bandwidth, but to say that this isn’t a “fast lane” is just hypocritical.
Furthermore, recent trends also show that private content networks are growing at a much faster rate than what we consider the “public” Internet, but this is a good thing because the massive amounts of digital content flowing over these private networks is not flooding the general purpose Internet. For Google to suggest that it should be illegal for network operators to build “fast lanes” and sell access to companies who can’t afford to build their own private backbones seems to be very anticompetitive and self serving.
Outlawing existing business models
Google has been a heavy proponent of banning the business practice of “Levying surcharges on content providers that are not their retail customers”. It’s not certain if new FCC rules would attempt to do this, but H.R. 3458 spells the following out and almost mirrors Google’s sentiments.
- to protect the right of consumers to access lawful content, run lawful applications, and use lawful services of their choice on the Internet;
- not impose a charge on any Internet content, service, or application provider to enable any lawful Internet content, application, or service to be offered, provided, or used through the provider’s service, beyond the end user charges associated with providing the service to such provider;
So while Google is says it’s wrong to charge content providers, Google is being hypocritical when it pays ISPs to host their edge caching servers which gives them the ultimate fast lane and the ultimate content distribution advantage.
More importantly, innovative devices like the Amazon Kindle rely on this business model. Amazon is not a “retail” customer of Sprint, yet Amazon pays Sprint to allow their Kindle users to get on the network at no charge to the user. Kindle users simply pay for the Kindle device and the books they choose and Amazon pays for network connectivity on both the server end as well as the consumer end. The catch is that Amazon operates a “walled garden” where Amazon chooses which websites and which blogs Amazon Kindle users get to access. Without this arrangement where Amazon pays for the “shipping”, Kindle users would have to pay $40 or $60 per month for full service mobile Internet access when all they need is the ability to download a few tiny books. While this goes against the idea that content providers shouldn’t be charged by ISPs for access to customers and that no one should control what users can access, it shows that innovation doesn’t always fit what Net Neutrality utopians envision.
Net Neutrality regulation also requires consumers to have the right to have access to all lawful content and services. However, that “right” should not mean an “entitlement” to have those things for free nor should it take away the right for consumers to have the choice to pay less for more limited services. Consumers still have plenty of choices to buy hundreds of devices and dozens of services that can access the full Internet but they can pay their own way. Other consumers who don’t need full Internet on a particular device or service should not have to pay for features they don’t need and they shouldn’t be forced to subsidize those features for other users.
There are devices and service plans which offer a wide range of features and capabilities, and many of those services do limit access to the Internet. In figure 3 below, we see devices and services that range from the fully featured and most expensive to the least featured and least expensive and it is all is all perfectly reasonable. Yet Net Neutrality regulation would outlaw most of these service plans even though consumers benefit from them. Consumers deserve the full spectrum of choices and not just the ones that Net Neutrality proponents see fit.
Figure 3: A la carte Internet access
Net Neutrality sounds great until we start looking into the details. While some of the proposed regulations and rules may sound reasonable on the surface, the architecture is far too complex to be to redesigned by non-engineers and the business models are far too diverse to be hammered into a single business model. The Internet has been a vibrant and innovative phenomenon that deserves the freedom to progress without the heavy and misguided hand of regulation.