Questions for Witnesses in Tomorrow’s Net Neutrality Hearing
Chairman Doyle is kicking off his term as chairman of the Communications & Technology Subcommittee in the House tomorrow (February 7). The witness list – and Doyle’s statements – suggest that he seeks to reinstate the Obama FCC’s Title II Open Internet Order from 2015.
This would be a mistake because the order is ambiguous, incomplete, poorly reasoned, and dependent on poorly drafted portions of the Communications Act. Any legislative review of net neutrality has to start with correcting the problems created by the 1996 revision of the Communications Act.
Primarily, the Act focuses on telephone service and is almost completely silent about the Internet. Since 1996, we’ve come to realize that the Internet has a problem with data protection. Correcting this problem will resolve consumer concerns about privacy and data security, and, if done correctly, will resolve the net neutrality issue as well.
Neither Title I nor Title II is a Good Tool for Regulating the Internet
Title II was intended to promote competition for local telephone service and to stimulate the creation of information services. It’s the end point of 30 years of discussion on the uses of and constraints on digital telephony.
This discussion started with the initial development of computer networks, long before the Internet was created. It arrived at a simplistic separation of computer services from the digital telephone network – the Title I/Title II distinction.
This separation made rudimentary sense before the Internet when we knew what a transmission service was and how such services differed from computer time-sharing and rudimentary database services. But today, it simply creates confusion.
The Internet Muddied the Title I/Title II Distinction
While the Communications Act deals with real networks and real services, the Internet is a virtual network. In the world imagined by the Act, there is one and only one transmission service. This service goes everywhere and connects everyone.
The Internet doesn’t work that way. It incorporates many transmission services (Ethernet, Wi-Fi, 4G/5G LTE, DSL, cable modem, xPON, et al.), none of which is pervasive. Some activists claim Internet Protocol (IP) data service is the new transmission service, but IP can’t actually do transmission by itself.
IP depends on data link layer systems such as Ethernet for transmission. So it’s a best a secondary transmission service. IP is also not simply provided by carriers; the largest IP data networks are owned by firms such as Amazon that don’t connect to consumer homes.
Transmission Services are Everywhere
The Act defines telecommunication as “transmission, between or among points specified by the user, of information of the user’s choosing, without change in the form or content of the information as sent and received.” According to this language, email, VoIP, Twitter, WhatsApp, Facebook, YouTube, and DVD rental are telecommunication.
In fact, Congress and the FCC have long debated whether VoIP and email are transmission or information services. We all probably understand why they aren’t telecommunication, but that requires us to put the statutory definition aside and go with a “I know it when I see it” approach.
In reality, the Internet performs transmission at several layers. It has a recursive architecture that depends on functions being performed over and over as information moves from point A to point B. We’ve discussed this in our podcasts and videocasts. We need a sound definition of telecommunication because everything that isn’t telecom is information service.
The most important question to be asked to the panel at tomorrow’s hearing is: “How should we define the core services that we want to regulate with net neutrality rules?” The answers will need to encompass definitions of the Internet, Internet Access Service, Internet Interconnection, and transmission.
Witnesses should also be able to define throttling, prioritization, paid prioritization, and reasonable network management. Without clarity on these questions no enforcement action will make any sense and no service provider will have any certainty about what they can and can’t do.
Witnesses should also be asked to comment on the relationship of the data transmission functions net neutrality seeks to regulate with the data management practices of Internet-based services such as Facebook, Twitter, and Google. The panel should seek to understand which is more important to Internet users, and how net neutrality regulations affect privacy, data security, and other data management practices.
Glaring Defects in the Title II Order
In addition to failing to define paid prioritization in anything like a meaningful way, the Title II order contained vast giveaways to Silicon Valley giants. It essentially made interconnection free, for example. This is less an issue of money changing hands than an abandonment of the obligation to coordinate network upgrades with affected parties.
Before Netflix reconfigures its network to shift data exchange from the One Wilshire data center to the Market Street center in San Jose, affected ISPs need to know so they can ensure they have the capacity to handle the load. When interconnection is governed by contracts, these discussions take place.
The Title II order also protected large Internet ad brokers (Google and Facebook) from competition from ISPs by altering opt-in/opt-out defaults. Is it reasonable to apply different presumptions to different players in the Internet ad business, and even different agencies (FTC vs. FCC) to enforce them?
One Big Rule to Replace Net Neutrality
Net neutrality is a failure because it doesn’t protect consumers from the most important risks and harms we face in using the Internet today. It does not preserve freedom of expression, it does not protect personal data from abuse, it does not limit surveillance, and it does not limit piracy and other forms of criminal conduct.
Moreover, it does not discourage the formation of monopolies or encourage the creation of new information services. It has failed because it did not anticipate the way the Internet services landscape would develop after 2003.
The problem we have today is the abuse of personal data. Blocking, throttling, and harmful (not simply “paid”) prioritization are best understood as basic forms of data abuse. So too are privacy violations, malware, unauthorized surveillance, misleading and manipulative advertising, and copyright violations. We need one big rule protecting data from abuse.
So my final question would be: “What suggestions can you offer for an Internet Data Protection Act that would apply to all parties doing business on the Internet? Please bear in mind as you answer the facts that advertising is essential to the financing of Internet-based services and that the processing of big data sets is essential to ensuring that consumers will reap the benefits of the Information Age.”
In other words, the time has come for Congress to stop playing divide-and-conquer games pitting one part of the Internet Economy against the other. We all need an Internet that is powerful, safe, robust, and dynamic.
Fundamentally, the practices that concern voters have nothing to do with the design of the law or the structure of the Internet’s internals. We want a system that allows us to do what we want, when we want it within the limits of human decency. We care about tangible effects, not about symbolism and historical trivia.
Net neutrality was created at a time when the only large firms conducting Internet business were ISPs. It was sensible for lawmakers to focus on ISPs in 2003. But today’s Internet is dominated by non-ISP edge services that routinely abuse personal information. Internet law needs to leap forward to the present day by addressing real issues.