Broadcast Media

Chapter 1: Net Neutrality as a Debate about More than Economics, by Christopher T. Marsden, from Net Neutrality:

Towards a Co-regulatory Solution, is available under a CreativeCommons Attribution-Non Commercial 3.0 Unported

license

1

Chapter 1. Net Neutrality as a Debate about More than

Economics

The topic today is net neutrality. The Internet today is an open platform where the

demand for websites and services dictates success. You‟ve got barriers to entry that are

low and equal for all comers … I can say what I want without censorship. I don‟t have to

pay a special charge. But the big telephone and cable companies want to change the

Internet as we know it. They say they want to create high-speed lanes on the Internet

and strike exclusive contractual arrangements with Internet content-providers for access

to those high-speed lanes. Those of us who can‟t pony up the cash for these high-

speed connections will be relegated to the slow lanes. So here‟s my view. We can‟t

have a situation in which the corporate duopoly dictates the future of the Internet and

that‟s why I‟m supporting what is called net neutrality. [1]

Barack H. Obama, 8 June 2006 podcast

This is a book about net neutrality. It is intended to be read by the non-technical as well

as the technical reader, by the non-economist as well as the economic, and also most

definitely by non-lawyers. It is most dense in its economic and legal analysis of

telecommunications in the opening chapters, yet those who are not telecoms/Internet [2]

policy „wonks‟ will find plenty to interest them in later chapters that focus on regulatory

agencies, consumer welfare more broadly, media policy and freedom of expression,

and political judgments such as the stated opinion of President Obama I have

highlighted above. It does not accept the neo-classical price-oriented competition-based

analysis which has been prevalent in telecoms policy for the past decade to 2008,

preferring instead to analyse from that base, to discover net neutrality to be a problem

of consumer and media policy. However, its critique of telecoms policy will provoke

(perhaps wry) interest in those still wedded to the concept of telecoms based on the

siege warfare over cost-based pricing between incumbent monopolies/oligopolies,

competitive market entrants and regulators.

It is an international book, in that it is written by an English lawyer who now preaches [3]

but used to practise, [4]

but with significant input from study of the European Union [5]

(and

European Economic Area) more widely, and with attention paid to North America [6]

and

Japan, [7]

if less so to developing countries. At the end of the book, I expect you to

disagree with me, whether you are a traffic prioritization free-market „RoundHead‟ or an

information-wants-to-be-free fundamentalist net neutrality „Cavalier‟. My argument will

be a „Middle Way‟ between these extreme positions that strikes a balance between

2

intervention and innovation, which inevitably means no-one will be happy, including me.

It is not a debate with any easy non-controversial answers.

What is net neutrality? Legalistic and technical definitions will be compared and

contrasted throughout the book as it is as much a term of art as a term of science. By

way of introduction, I should lay out what it is not: it is not a panacea in the absence of

effective inset competition in telecoms markets. In the United States, urban duopoly and

rural monopoly telecoms throughout the latter Bush years led to significant fears of

gatekeeper control over information flows by Internet Service Providers (ISPs).

However, net neutrality is an issue that arises in all competitive and non-competitive

information environments that use Internet Protocol (IP) and the public Internet to

communicate. World Wide Web (WWW) inventor Sir Tim Berners-Lee puts the problem

like this:

Net neutrality is this: If I pay to connect to the Net with a certain quality of service, and

you pay to connect with that or greater quality of service, then we can communicate at

that level. That‟s all. It‟s up to the ISPs to make sure they interoperate so that that

happens. Net Neutrality is NOT asking for the Internet for free. Net Neutrality is NOT

saying that one shouldn‟t pay more money for high quality of service. We always have,

and we always will. There have been suggestions that we don‟t need legislation

because we haven‟t had it. These are nonsense, because in fact we have had net

neutrality in the past – it is only recently that real explicit threats have occurred. [8]

In short, net neutrality is about the rules of the road for Internet users, and about the

relationship between the owners of those roads and the users. Government is asked to

make a decision as to which users have priority and whether road charging should be

introduced, ostensibly to build wider and faster roads in future. [9]

That is a profound

issue, and that brief summary tells you there is plenty of critical detail within this

argument that can shift the balance of advantage towards networks or users, and

between different players within those groups.

Let‟s therefore crudely summarize the argument using the slogan that became the

rallying cry for Generation X a quarter-century ago: „I want my MTV (Music Television).‟

Users who wish to access video and other high bandwidth content cannot all do so at

once over current networks, and probably not over future networks. Currently there is no

speed limit for the Internet, nor are there sufficient „lanes‟ for all users to travel at the

speeds they wish (assuming their „car‟, their Personal Computer or other device, can do

so). So if you want your British Broadcasting Corporation(BBC) iPlayer streamed video,

or Skype video call, or YouTube clip, or World of Warcraft update, you are in a queue,

even if you don‟t realize it. In a future with guaranteed Quality of Service (QoS) – a

3

future that has been waiting for 30 years and launched a thousand doctoral theses in

computer science and network engineering – you may be able to pay more to get to the

head of the queue, or the content provider may pay more to get to you faster, past the

other members of the queue whose content providers won‟t pay. You may also be lucky

enough to get into a queue that moves much more quickly for everyone (such as the

university-funded Internet2), or unlucky enough to have to access your content over the

slow-or-never mobile or wireless options, [10]

which at least offers mobility.

So why is this net neutrality „problem‟ arising now? Two reasons. First, in developing

countries most Internet users – if not most people – have broadband connections. But it

turns out that those broadband connections are not fast enough for the uses that people

want – not just YouTube but also public service content, for instance BBC video

programming or online tax returns, as well as Peer-to-Peer (P2P) content. People are

upset, particularly when they realize that they are sharing a connection with their

neighbours, and that their speeds slow down dramatically in „rush-hour‟ – the evening

when they sit down to watch video on their computer. They bought connections that

advertise „maximum speeds‟ (i.e. unlimited by neighbours‟ congestion and other speed

bumps on the Internet) of perhaps 8 megabits per second (8 Mbps) – enough to

download 600 megabytes [11]

of data in ten minutes, or a five-minute video clip of

reasonably high quality. That speed means they can click and almost immediately start

watching the clip, and it will keep downloading in the background as fast as they are

watching – so that they receive a seamless video clip. Downgrade the quality of the

video and you can download the quality of the speed – but as a user, you don‟t want to

go back to the future!

This now gets tricky – the speeds need to increase in line with customers‟ expectations,

especially as most of your neighbours are now online – at the same time as you. Your

expectations as a consumer are in part developed from your business or especially your

university broadband experience, where you will probably have much faster speeds that

enable you to download and upload much faster, and share content such as slide

presentations by email almost instantaneously. Your office will have fibre-optic

connections – that‟s basically a motorway for the office – which solves most of the „last

metre‟ problems, at least. Your home won‟t, not yet and perhaps not ever. It will have a

copper pair of wires that were designed for circuit switched telephone calls, down which

the phone company squeezes the data packets using technologies based on Digital

Subscriber Line (DSL) standards. If you have cable, you have a similar set of problems.

Your connection is slow, shared and uses a physical link that is obsolete. But count

yourself lucky, if you use mobile or wireless, you could wait forever for peak-time

content.

4

A Brief History of Network Neutrality

Ten years ago, 4 June 1999, I hosted a conference „near Stratford Upon Avon‟ in

England, inviting some US Internet lawyers to speak to UK communications specialists

about the new discipline they were forging, the possibilities it offered and the dangers to

its progress. I launched a book with my own and my friends‟ views (British, European

and American) on the subject from a European perspective, Convergence in European

Digital TV Regulation, warning that the Internet unregulated model would have to

accommodate the normative democratic values of the regulated European public

service broadcasting model on the one hand, and the economic imperatives of e-

commerce on the other. In the book, we also warned that the Information Society would

depend crucially on broadband infrastructure, a theme returned to in the book of the

conference, published in autumn 2000. A rash of works by conference attendees

appeared in 1998–9 on the same theme such as Lemley and Lessig, [12]

including path-

finding works by Froomkin, Reidenberg and Samuelson. [13]

One particular issue for Lessig and Lemley that summer was the merger of long-

distance phone company AT&T and cable company MediaOne. They told the FCC in an

ex parte submission to the merger investigation [14]

that they feared the „end of end-to-

end‟. [15]

In brief, they feared that AT&T‟s Internet openness, in as much as it existed,

would be subsumed into the closed pay-per-view world of cable TV companies. AOL,

the original „walled garden‟ Internet provider (offering filtered, approved and partnered

content in a „safe‟ environment, as well as open Internet access for the brave), was

merging with Time Warner, a copyright and cable behemoth, which threatened an even

more closed experience. [16]

The interoperability debate is broader than simply an

Internet access debate, as it affects innovation in software – indeed, the origin of the

argument lies with software industry disputes over interoperability, an argument

captured by Lessig in his contribution to the Microsoft litigation. [17]

In spring 1999, P2P music sharing site Napster was sweeping college campuses, and

the music and movie companies were crying out for protection from the evils

perpetrated against their cozy copyright oligopolies by this new P2P technology for file-

sharing. I was interested in their argument, and in file-sharing as a technique to break

open the broadcast/music oligopoly, though I argued that the European, or at least

United Kingdom, relevance was slight as we had almost bankrupt US-owned cable

companies and a telecoms monopoly that was denying that demand for broadband

existed. In brief, we should be so lucky as to have two broadband companies that

threatened to take a high-handed approach to their users‟ access. I claimed, following

Noam and others: [18]

5

The answer increasingly employed is to use the guaranteed service quality and

enhanced security of the „walled garden‟ broadband service providers‟ network, to avoid

the public Internet altogether. These „walled gardens‟ have a very satisfactory legal

status: they are cable networks. The private network ensures integrity of rights, video

delivery, and allocation of property. … The legal framework will ensure that this

broadband VOD [Video On Demand], when it arrives, will be more the AOL-style „walled

garden‟ than true open access: private cable not public Internet.

The lack of legal certainty in assigning property rights, whether to one‟s personal

information or spectrum for 3G mobiles and local loop unbundling (LLU – sharing the

copper wire from telephone exchange to subscriber), was restricting the growth of a

broadband Internet, and leading to a localized, Balkanized „walled garden‟ private

network approach, as Coasean analysis would suggest. [19]

In such a fragmented future,

the issue of open access to those private networks is critical. Without a more legally

certain international allocation of property rights, the old national legal restrictions will

continue to apply to profitable mainstream operators, and I wistfully remarked that „the

public Internet [will be] a source of piracy, romance and buccaneering on the high seas

beyond the reach of national legal certainties‟. This was not just because of the law and

economics of the networks, but also the requirements imposed by governments and

copyright and privacy lawyers on the content.

If you are a European telecoms lawyer, you will already know the quick march through

policy that we need at this point [20]

to understand where we have got to, but both US

and European observers need to know where their policies diverged in 2001 to predict

whether we are converging in 2009. I continue this analysis more extensively in

Chapters 1 and 2, but here‟s the summary.

European Telecoms Liberalization

The EU framework for telecommunications consists of the sector-specific ONP („Open

Network Provision‟) framework operated by the European Commission and the National

Regulatory Authorities (NRAs) of the Member States of the European Union, and EU

competition law applied by the European Commission and the national competition

authorities and the court system. [21]

Corresponding authorities in the United States are

the FCC and the US antitrust system. [22]

The EU telecommunications sector was

liberalized by three legal means:

 Liberalization directives issued under EU competition law, Article 86 of the Treaty

Establishing the European Community

6

 Harmonization Directives issued by the European Parliament and the Council on

the basis of Article 95 of the Treaty Establishing the European Community

 A series of merger cases in which the dominance of horizontally and/or vertically

integrated operators was structurally curtailed, by divestiture of cable

networks, [23]

competing mobile networks [24]

or even Internet backbone

networks. [25]

The measures culminated in the adoption by the European Commission in 1996 of the

Full Competition Directive, [26]

mandating full liberalization of telecommunications in the

European Union in 1998, and a flurry of merger cases between 1998 and 2000. 1996

was also the year that the United States broke open its regional monopolies via the

Telecommunications Act 1996. However, these measures left in place the incumbents‟

dominance of the local loop, the „first hundred feet‟. [27]

It is economically unfeasible for a

telecoms operator to duplicate that local loop, though urban cable networks have done

so in several countries (notably North America, Germany, the Netherlands and

Belgium). Therefore, the telecoms sector has been regulated using an extension of the

„essential facilities‟ doctrine [28]

to the local loop and other monopoly infrastructure, to the

chagrin of economists who believe that the North American cable build-out could have

presaged similar competitive build in other countries, had shared access to the local

loop been denied to them. [29]

Nevertheless the European Commission (EC) opened the

loop to competition in 2000, or rather to each national regulator‟s interpretation of how

competitive it wanted its local loop to be, a process that is still ongoing. [30]

The progress

of European telecoms liberalization over twenty years has been slower than hoped for

in most cases, [31]

despite which incumbents have rolled out DSL to the vast majority of

their populations.

The EC at the same time was engaged in its protracted action against the vertically

integrated monopoly of Microsoft over computer operating systems, servers and media

player software, which was settled in the United States during 2000–2 [32]

at the outset of

the Republican neo-liberal period in the United States, but continued to rumble on in

European courts. [33]

I should make it clear that the widespread European view is that the

George W. Bush presidency was an aberration in its adoption of neo-liberal agendas

not only in foreign imperialist policy but also in competition policy, supported for the first

six years by a compliant Republican House and Senate, which were under Republican

control from 1994 to 2006. Therefore, the rolling-back of competition in US telecoms

markets in 2001–6 can be seen from a European viewpoint as a historical anomaly, a

period in which neo-liberal economic models succeeded over much empirical data

demonstrating continued market failure and abusive monopoly. [34]

7

The EC was not especially brave in its decisions made after full liberalization in 1998, in

part out of deference to member state NRAs which were beginning to operate the

system and were given a certain amount of latitude by the EC, latitude that is rapidly

decreasing as it becomes apparent that almost 12 years after full liberalization, the now

27 NRAs have hugely divergent effectiveness and regulatory commitment to

liberalization. Back in 2000, architect of liberalization Herbert Ungerer felt confident

enough to state: „The approach of close cooperation with national regulators turned out

to be largely successful.‟ [35]

The pricing of interconnection and mobile termination were

two issues that continue to reoccur despite being temporarily „solved‟ in 1998–9. [36]

Ungerer stated: „Recent regulatory and antitrust decisions tend to be a mix of structural

and behavioural measures.‟ [37]

By 2009, the EC was pushing Member States to adopt a

supranational authority – Group of European Regulators in Telecoms (GERT) [38]

– that

would be somewhat similar to the US FCC and ensure conformity by NRAs.

Unsurprisingly, those plans had been watered down by governments who set up and

supported the more recalcitrant NRAs. The European Commissioner, Europe‟s premier

political figure in communications policy, in 2008 addressed incumbent telecoms

operators (telcos) on progress: [39]

Ten years after the opening of markets to competition, the job of regulators is only half

done. I know that you do not like me saying this. But this is my role as European

Telecoms Commissioner. It is because of your economic and political power that the

Commission has to remain vigilant, as the independent guardian of competition in the

European Union … I know that it may be convenient, in the short term, to enjoy the

protection of national rules and regulations … Perhaps, this will allow you for some time

to keep competition from abroad at bay. And to prolong badly needed transformation

and modernization processes a bit longer. [40]

However, back in 2000, it appeared that Europe was following the United States which,

through the decisions in AOL/TimeWarner and AT&T/MediaOne, was really moving

towards a „new Kingsbury commitment‟ in cable, referring to the 1913 agreement that

permitted AT&T a monopoly subject to universal service. The mergers of 2000 opened

the cable networks to the theoretical promise of competition, which brought them into

step with the Telecommunications Act 1996 provisions for LLU (known in the United

States as „unbundled network elements‟ or UNE), and enforcing competition in Internet

backbone services. It came as something of a rude shock when these commitments

unravelled at a furious pace in 2001–2, as the Internet bubble burst and analysts such

as Lessig and Wu [41]

realized that policy needed to rally round a principle: that principle

was network neutrality. Whether net neutrality is the new common carriage, as Sandvig

suggests [42]

and I unpack in Chapter 1, it is an idea of genuine descriptive power.

8

Bubbles Burst, Incumbents Delay, Deny, Degrade

Back in Europe, a year passed, the dot-com bubble that was so shiny, soapy and

opaque imploded dramatically in summer 2000, the EU regulators also imposed

openness conditions on AOL‟s merger with Time Warner, the largest in history, by

requiring divestiture of its part-ownership of a large German ISP. Europe auctioned its

3G wireless licences in a messy, highly expensive process, costing hundreds of billions

of dollars at the height of the bubble (paid after it burst), which further delayed any sign

of broadband Internet access for the many. With the dearth of high bandwidth content

start-ups (a self-fulfilled prophecy in the vicious cycle of capital-starved start-ups and

bandwidth-starved consumers), British Telecom (BT) and other incumbent telecoms

operators (telcos) continued to deny broadband demand existed, and new competitors

by late 2002 were in terrible financial trouble in the wake of the Enron, Global Crossing

and WorldCom scandals. [43]

Napster was bought by Bertelsmann Music Group but was

then sued into submission, replaced by more powerful, often encrypted and less legally

suspect technologies such as Gnutella, KaZaA, BitTorrent and others. I crawled out of

the wreckage of MCI WorldCom, whose CEO and CFO had committed the largest

accounting scandal in the world to that date, a staggering $11 billion (peanuts compared

to the Citicorps and AIGs and all the others in 2007–9, of course). The broadband

model was bust, I declared to the few who cared, and it was time for a Keynesian-

Korean stimulus to mend it: to prime demand by fusing John Maynard Keynes‟ insights

into recessionary spending on public works with the highly effective government-

directed but ruthlessly competitive Korean environment for broadband deployment. It

did happen in Japan and Hong Kong, which fast outstripped Korea in high-speed

broadband by using fibre-optic cables into the building to achieve real broadband

speeds of 100 Mbps. Meanwhile, BT in the United Kingdom and its European

neighbours finally rolled out broadband in 2002–4, and Britain by 2008 had a dizzying

2–8 Mbps „theoretical maximum‟ broadband connections for consumers. A few outliers

delivered broadband more quickly, where viable cable and telecoms networks belonged

to rival groups, [44]

but these were very few and confined to the Netherlands, northwest

Italy and Scandinavia.

There was another piece to the „end of end-to-end‟ story: the content delivery to the

consumer. Could this be achieved using the open Internet, where video would be

delivered across continents at the same speed and QoS as all other bits, whether spam

or P2P? Or would it need caching close to the end-user and the „last mile‟, to ensure

higher quality and take traffic off the network? I thought the latter, and further thought

that Hollywood would be encouraged to put its wares on the Internet if this could be

achieved. I wrote that we didn‟t have any type of broadband Internet yet in Europe, and

that to achieve it we needed first to host content locally. Between drafting this „Start of

9

End-to-End‟ paper in 2001 and finally publishing it in 2003, I rethought my views but

came to the same conclusion: „Lord give me openness, but don‟t give it yet‟ – we

needed an interim step on the way to a truly effective broadband Internet.

The European incumbent telcos had paused in their rush to the cable model, in part

because of those regulatory fears of unbundling, of being forced to share their

broadband networks with competitors, but largely because they did not have either

content or bandwidth to make it worthwhile. Voice phone calls are simply bits on the

Internet if there is sufficient bandwidth, and phone companies faced bankruptcy or at

the least „commoditization‟ if they permitted users to simply buy broadband and use it

for Voice over Internet Protocol (VoIP). This was not a new fear by any means, and was

the stuff of popular wisdom in 1997 thanks to „The Death of Distance‟ by Frances

Cairncross. [45]

The easy way out for telecoms companies was to sit on their hands and

employ an army of economists and lawyers to convince the regulator that unbundling

was inefficient and/or morally wrong and sinful in a „free market‟. Note that telecoms has

never been a free market but a cabal of monopolists with a fringe of competition from

the astute, regulatory minded and – in WorldCom and Enron‟s cases at least – corrupt.

Having sat on their hands, the phone companies found that most problems magically

simply melted away, as if in some Taoist fairytale – their competitors were bankrupt,

most content companies were happy to maintain a non-Internet status quo, the

regulators were impotent in the face of the highly paid economist hired hands and the

phone companies were home free. They forced consumers to buy phone lines with their

broadband access, [46]

kept prices high and blocked or degraded content they didn‟t like

using snazzy new filtering equipment called Deep Packet Inspection (DPI) using blade

servers and other ultra-high-speed computing power. As Riley and Scott explain:

In the early days of the Internet, non-discrimination was easy to uphold because it was

not technologically feasible for service providers to inspect messages and evaluate their

content in real time. But recently, electronics manufacturers have developed so-called

DPI technology capable of tracking Internet communications in real time, monitoring the

content, and deciding which messages or applications will get through the fastest. [47]

In short, telcos thought they had the field to themselves, except for a nasty-looking start-

up in 2002–3 called Sky P2P, Skype for short, by that annoying Swede Niklas

Zennstrom, who had dreamt up KaZaA in 2001. However, given that it was a P2P

based Instant Message service masquerading as voice, they thought they could block it

wherever it formed a serious threat. They had originally lobbied to block it from

advertising itself as a phone service as it does not work in emergencies when the power

cuts out and your computer battery dies (after all it is a cheap substitute with limits). By

2004, ISPs were being caught blocking a rival to Skype called Vonage in the United

10

States. The other perennial problem for phone companies was the lobbying by the

copyright industry to get ISPs to cough up the names of their subscribers who were

sharing lots of P2P files (KaZaA had by now replaced Napster and soon was joined by

Grokster and BitTorrent and other more advanced P2P technologies). Copyright players

had decided since the dot-com bubble burst that their existing business model was just

fine, thank you, and they were not forced to return to the table until Steve Jobs and

Apple convinced them to join the iTunes digital store – by which time literally billions of

copyrighted files were being swapped by users who did not want to return to that

twentieth century business model of buying hard discs. ISPs resisted phone companies‟

attempt to get them to join in the war on their customers, heroically in the case of

Verizon‟s battle with the recording industry. [48]

In desperation, consumer advocates sought a light at the end of this dark 2002–4

tunnel. They found it in what became known as „network neutrality‟, the Lessig–Lemley

principle of „end-to-end‟ (E2E), based on an engineering principle that the early Internet

worked best when all packets were routed with the same priority. Internet engineers had

been working to refine the network for years, concerned that the „one-size-fits-all‟ E2E

approach was not sufficient for high quality and/or large volume transactions, in

particular those Holy Grails voice calls and video transmission. In 2002, Tim Wu

described „net neutrality‟ in a working paper that became a 2003 ex parte submission to

the FCC jointly with Lessig. Battle became joined at that point, with the cable and telcos

claiming that such was their duopolistic competition, and so near the real competition

from the third wireless broadband alternative, that they needed no such regulation. If

they were going to fiddle with a customer‟s Internet, they warned, they would only do so

in properly utilitarian manner for their benefit, or at least that which would bring greatest

happiness to the greatest number of users, stockholders, bondholders, executives and

the like. Companies are not governments, so their concern had to extend across these

various constituencies, and profitable investments had to come first in that list of

priorities.

That wireless alternative was becoming more than just theoretically interesting. The

farcical European 3G auction may have delayed the build-out and aggressive marketing

of 3G in Europe, which had led the United States in voice wireless through the 1990s,

but 3G was very real in Japan by 2002. It was however almost the polar opposite of the

open Internet pursued by net neutrality advocates. It was in fact a „walled garden‟,

inside which were the preferred partners of the network operators, of which by far the

largest was NTT DoCoMo, the wireless arm of the incumbent. It charged its preferred

partners 8% transaction fee on their sales, and provided for part of this tariff its billing

services to the end-user. Thus ringtones, e-commerce sales and so on would appear on

your wireless bill, truly a „one stop shop‟. To go from this „walled garden‟ into the wider

11

Internet took you away from the trusted partners of DoCoMo and into a slower and less

secure environment.

That sounds like a fairly claustrophobic experience for more adventurous users, but an

alternative had presented itself: Wireless Fidelity (WiFi). This was an in-building

unlicensed spectrum technology that was simply supposed to enable the end-users to

work wirelessly inside their own premises. However, boosting power, marketing the

trademark „WiFi‟ standards and building the technology into the laptop computer while

mass producing wireless modems made the obscure Radio Local Access Network

(RLAN) into a device owned by hundreds of millions by the middle of the decade. As I

wrote with Ivan Croxford in 2001, „I want my WiFi‟ was the consumer call of the time

(echoing „I want my MTV‟ of the cable television generation). Moreover, consumers

showed demand for national „hotspot‟ coverage of WiFi, first in obvious transport hubs

like airports and railway stations, then in „road warrior‟ salesman refreshment stops like

Starbucks. ISPs such as T-Mobile and Boingo and start-up TheCloud supported by BT

aggressively partnered with such locations to provide what looked like national urban

coverage in the tens and hundreds of thousands – this despite the RLAN being illegal

for outdoor commercial use until 2002 in many countries. [49]

None of these networks

made much money, though they supplemented existing ISP wired and wireless

networks, and it is now the case that dozens if not hundreds of WiFi hotspots are

available in most urban locations (check your network availability as you read this

book).

This phenomenal growth of WiFi occurred largely against the wishes, contractual terms

and laws of their hosts, as Wu explained in the original net neutrality paper:

Operators showed an unfortunate tendency to want to ban new or emerging

applications or network attachments, like WiFi devices or virtual private networks,

perhaps out of suspicion or an (often futile) interest in price discrimination. [50]

Why? Because it was free, it could be shared with anyone who had the password or

could access the unprotected hotspot of a neighbour, and therefore it threatened to

„cannibalize‟ the revenues of ISPs. If consumers could achieve the theoretical

maximums of 56Mbps on their WiFi, they could share data and voice with dozens of

their „friends‟. Several legal obstacles were rapidly constructed to stop consumers doing

this: first, they were in breach of their contract with many ISPs if they shared – even

unwittingly – their wired Internet connection outside their own household. Second, they

could be held criminally liable if someone used their network to download illegal matter

such as child pornography as shown in a Canadian test case. Third, they could be held

civilly liable if someone used their network to, for instance, download copyright material.

Importantly, these latter pair of legal problems came equipped with a standard-based

solution: the encryption on WiFi base stations was made more powerful and more user-

12

friendly. As a result, most users closed their networks to strangers and prevented a

spectrum commons of WiFi free hotspots. However, in some locations, notably

Montreal, Canada, the „Isle sans Fils‟ (Island without charge), social entrepreneurs have

established free Internet networks using WiFi.

The backdrop to the net neutrality debate is therefore a series of revolutionary

technological and user-centred breakthroughs, with the unmetered Internet itself only a

decade old for European consumers, broadband only 5–7 years old, WiFi the same,

Skype and other VoIP even younger. File-sharing via P2P dates to Napster in 1999–

2000, with KaZAa and BitTorrent even younger. Broadband providers faced with this

tsunami of innovation reacted quite logically in trying to slow down the pace of change,

first by delaying the introduction of broadband, then by slowing its spread via WiFi, and

by throttling P2P applications. Video and movie providers shelved previous plans to

rapidly deploy Internet-based video in the „dot-com‟ bubble of 2000–1, going into

partnerships with broadband providers and P2P technology companies. The growth of

bandwidth was choked off while both video producers and ISPs puzzled over a way to

„monetize‟ – to profit from – the new delivery technology.

In summary, telcos are trying to avoid a commoditization of their business, which means

they fear and envy their mobile counterparts. Fear – because the user-friendliness and

personalization of mobile phones means users are increasingly relinquishing fixed lines

for mobiles, as well as investing heavily in more sophisticated terminals and expensive

monthly subscriptions. Envy – because mobiles have been able to persuade users to

pay for usage and speed of service, and content partners to pay for access to a higher

quality, faster and more „trusted‟ portal. In Japan by the end of 2008, almost 90% of

mobile users had broadband services, and telco fixed-line services have been

overtaken by fibre-to-the-home (FTTH) (13.1 m to 12.3 m with 3.9 m cable broadband).

There were five mobile subscribers for every two fixed subscribers (105 m to 43 m). In

broadband, the 90 m mobile users outnumber DSL users by 7 to 1 and total fixed

broadband users 3 to 1. [51]

Even worse for fixed companies, VoIP subscribers reached

30% of fixed (10 m users). VoIP users typically pay far lower per minute charges and

are therefore much less profitable than dedicated phone subscribers. VoIP is not

permitted by most 3G mobile networks, which preserves their telephony revenues. In

Austria, an even more potentially devastating substitution has taken place, with 3G

subscribers choosing to use 3G modems („dongles‟) on laptops and netbooks (mini-

laptops), and in 2009 that number is expected to exceed fixed broadband totals.

13

From History to the Future

That breakneck dash through recent telecoms history was necessary to give a little

context to the net neutrality debate in 2009. You have heard the term and have some

sense that it is important or controversial – or at least you must be working for a

company or taking a course where someone thinks it is important. It is important to

realize that (a) it is not a new debate; (b) it is not going away; (c) it grows in importance

as the Internet and the importance of users accessing high-speed content grows in

importance. So you‟ve invested wisely so far.

In Chapter 1, I will explore how companies and governments are trying to upgrade

home (and therefore small business) connections to fibre, or a close alternative called

Very high-speed DSL (VDSL). For now, let‟s focus on what is happening on your line

during peak-time, and what the phone company is going to do about it. Europe is only

just over a decade into unmetered Internet access for dial-up connections and it would

be a pity to go back to that type of future. So what‟s changed to threaten Berners-Lee‟s

open Internet? DPI reshapes strategy for traffic management so that what was formerly

a dumb network processing packets on a „best efforts‟ basis, can now configure all

kinds of re- and non-prioritization. Ridley and Scott state that:

Operators can tag packets for fast-lane or slow-lane treatment – or block the packets

altogether – based on what they contain or which application sent them … When a

network provider chooses to install DPI equipment, that provider knowingly arms itself

with the capacity to monitor and monetize the Internet in ways that threaten to destroy

Net Neutrality and the essential open nature of the Internet. [52]

The genie is very much out of the bottle regarding DPI and ways to make providers pay

for higher quality: it exists and is used by most incumbent ISPs and many competitors.

That means there is a very real issue about how this technology should be used. That

affects more than simply speed of access to content. It also affects the rights of the end-

user to receive content, and the privacy thereof, as well as the rights of content

producers to provide applications and services that end-users can access in reasonably

conducive conditions. A different book could – and must – be written about the uses and

abuses of DPI and other technologies, but for our purposes, we should accept that ISPs

can manipulate the bits on the Internet for their (and sometimes) our benefit. Should we

worry? We should trust but verify, at the very least.

All network owners have incentives to stop traffic flowing over their networks that is low

value, high volume and for which it is technically unfeasible or uneconomic to charge –

notably non-network affiliated content including user-generated and transmitted content.

14

This content is very low value to the network and, with many millions of users all valuing

each others‟ own-created Web 2.0 content, under current market and technological

conditions there is insufficient value to charge individual users and thus all content may

be throttled in the absence of a charging mechanism. [53]

Content on limited bandwidth

networks can „choke‟ the network capacity, especially at peak times of usage (daytime

for business, evening for consumers). In a „best effort‟ environment without congestion

charging, [54]

this content has insufficient disincentives to prevent its flourishing: for

instance P2P traffic and its use by early-adopter high-volume users. ISPs can choose to

filter P2P traffic of various kinds – typically it is unencrypted relatively crude versions of

popular file-sharing programmes, such as BitTorrent which is used to provide upgrades

to the most popular multiplayer online game World of Warcraft. Many assertions are

made about the implications of certain types of traffic, but regulators currently have no

basis for deciding if such assertions represent real problems. [55]

The following types of discrimination might constitute the type of non-neutral behaviour

by ISPs that may be found to be harmful to consumer welfare: transparency failures and

misleading advertising, „throttling‟ or blocking, charging, certain types of more extreme

and anti-competitive „walled gardens‟. [56]

First, transparency failures. ISPs may fail to tell customers and application developers

which services they offer – estimated bandwidth, latency, etc. This is essential to certain

applications, which cannot run with latency, or which are blocked or filtered. Even where

there is regulatory commitment to enforce net neutrality, the evidential problem remains.

Van Schewick [57]

suggested that the main problems currently lie in mobile networks,

where VoIP is routinely degraded or blocked. [58]

The problem here is that certain users

are breaching their terms of use but being insufficiently or non-transparently sanctioned,

and certain programmes are being throttled but the same applies. Often a security

justification [59]

is used and is often unchallenged by regulators.

Next, blocking or „throttling‟ is the furthest deviation from neutrality. Some economists

think it justified, but the basic problem is a distortion of competition between the blocked

and unblocked companies. For example, a company serving online gaming content

from South Korea may typically choose to do so via P2P networks, whereas a US

content provider might use a premium service sanctioned by the ISP of the end-user.

Not only is the Korean content provider discriminated against, but neither end-user nor

content provider may be aware of the nature of the problem. [60]

This creates confusion

among users as to whether and how content is throttled. [61]

Certain types of traffic that are highly valued by the end-user of the Internet can be

discriminated against in whole or in part by service providers that are not dominant. This

15

is because they either have good competitive or good traffic management reasons to do

so; it makes their networks safer and more efficient, making it complicated to work out

when their discrimination is motivated by arguably less benevolent factors, like blocking

the competition. There can be motives to throttle content no matter what ISP is

discussed, and that behaviour is potentially anti-competitive not within the layer of ISPs,

but to content providers upstream which end-users are trying to access via the ISP

network.

Blocking together with other forms of traffic shaping is particularly controversial

because, under current network management tools, it is a blunt tool. For instance, all

P2P traffic using a certain protocol may be blocked. P2P can respond by encrypting its

traffic or otherwise spoofing, but this creates an „arms race‟ much like that found in

security software responses to the threat of breaches. In fact, the claims of ISPs are

that P2P traffic contains a high proportion of malware, spam and spyware, and therefore

it is filtered in the end-user‟s interest and in conformity with the terms of use for end-

users. [62]

Many assertions are made about the implications of certain types of traffic, but

regulators have no basis for deciding if such assertions represent big or small problems.

The ISP assertion that P2P traffic contains a high proportion of malware may be

disingenuous. Email spam and web surfing are the vectors for malware, but the ISPs do

not block such traffic. Future networks may try to cap P2P more effectively, which can

itself lead to an „arms race‟ between encrypted P2P content and attempts by ISPs to

detect P2P traffic. This is an example of how a baseline of traffic and usage would help

the regulator to understand the importance of claims made by stakeholders. [63]

Since broadband ISPs have a termination monopoly or duopoly [64]

over the end-user,

they can use that to charge termination fees to those who wish to get access to the

user. This behaviour is familiar to the cable TV industry, where only large content

providers can secure free or even profitable carriage, whereas smaller content providers

with less contracting power are forced to pay the cable TV operator for access. The fear

is that a similar model will be imposed on the Internet, where only large content

providers with sufficient negotiating power, and those with political influence to secure

favourable carriage terms, will secure free carriage.

Next, let‟s consider P2P.The claim made is that networks cannot be upgraded

successfully given the flood of P2P traffic. This is by no means a universally shared

sentiment amongst ISPs and I note recent comments attributed to Matt Beal, BT

Wholesale‟s chief technical officer: „It is up to us at the core of the network to make sure

there is enough bandwidth‟. [65]

He further stated BT‟s Next Generation Network (NGN –

its all-IP network) [66]

would „put enough [bandwidth] volume out there … so we don‟t

have to [traffic shape]‟ which is „quite Big Brother-ish‟. There is therefore no consensus

16

as to the type and extent of traffic shaping and other forms of blocking and throttling

P2P traffic. Where ISPs do not have effective terms of use, or do not enforce uniformly

those current strategies in place to dissuade „unfair‟ use, two consequences can follow.

1. Users are summarily terminated or suspended – this can be conducted by any

ISP and may well be justified. This practice could be made more transparent. [67]

2. ISPs choose to filter P2P traffic – typically popular file-sharing programmes.

The rights perspective should be put openly. Davies and Banks explained: [68]

One of the most important trends in recent years is the growth of multinational corporate

censors whose agendas are very different from those of governments. It is arguable that

in the first decade of the 21st century, corporations will rival governments in threatening

Internet freedoms. Some American cable companies seek to turn the Internet into a

controlled distribution medium like TV and radio, and are putting in place the necessary

technological changes to the Internet‟s infrastructure to do so.

Increasing use of DPI is being created for both Western ISPs and more autocratic

governments. [69]

In both cases, the method chosen is co-regulation – the government

sets the rules and the ISPs are allowed a broad measure of independence as to

process to achieve the results the government sets out. This is controversial in that it

passes powers to control freedom of expression into private hands, often without the

constitutional protections that govern public authority intervention and censorship.

Whether it is China or the United States or Europe that implements the policy, the ISPs

are using similar tools and techniques and the language used in regulation is very

similar. [70]

The effects on user rights may be varied, but the techniques used are

undoubtedly filtering, a form of censorship.

This makes network neutrality more than an empirical positivist economic issue: it is

explicitly normative and political.

More Than Economics

Open Internet Policy

The network neutrality debate is only in part about economics and technology, despite

what you might surmise from various pro-competitive statements by academics and the

shape of the US and European debates. The extent to which even lawyers have been

drawn into an open-ended debate regarding the merits of duopoly versus inset

competition in telecoms, or the relative merits of open interoperable software

environments versus proprietary property rights-based or corporate developments, or

17

the benefits of end-to-end „dumb‟ networks versus intelligent networks, displays the

capture of the subject by economists and corporate technologists. The issues at stake

are more fundamental to society than that. As a lawyer who has written for over a

decade in favour of pro-competitive telecoms and media policy, I am not ashamed or

abashed to state that I emphasize that communications policy is about fundamental

rights of citizens as well as public welfare for consumers, and that it is about educated

and informed users as well as optimally priced access networks. In short, what is

needed is a balanced approach towards network neutrality as a central plank of a

converged communications policy, ideally one which tries to both increase competitive

choices for consumers as well as ensure the fundamental right for citizens to access the

public Internet.

This conclusion should not be startling to anyone who has studied communications

policy over the decades, though the capture of the debate by economists and

technologists for the benefit of network investment over open access is a long-term

trend that may not be reversed in the short term. I argue that it needs tempering with

considerations of fundamental rights, not overturning in favour of a dirigiste

broadcasting policy. I take my cue from several experts, noting that they are pro-market

advocates who also recognize a value to openness – that is, as opposed to private

censorship – which extends beyond economically quantifiable results. Lessig offers

caution as to the empirical outreach of economic analysis of law, explaining that its aim

is totalizing. This analysis reflects the fear of that totalization, the need for a „balanced

diet‟ between law and economics, [71]

and the influence and paramount importance of

human rights in policymaking, particularly here in the case of the rights to freedom of

expression and privacy. Rights may have been considered „nonsense on stilts‟ by

utilitarians two hundred years ago, but we have earned the right to consider them today

in the case of the Internet in developed countries, at the very least. Berners-Lee

explained that the open standard of the WWW describes:

a vision encompassing the decentralized, organic growth of ideas, technology, and

society. The vision I have for the Web is about anything being potentially connected with

anything. It is a vision that provides us with new freedom, and allows us to grow faster

than we ever could when we were fettered by the hierarchical classification systems into

which we bound ourselves. [72]

Lessig explains what the Internet architectural principle of „end-to-end‟ [73]

means for

innovation: [74]

This end-to-end design frees innovation from the past. It‟s an architecture that makes it

hard for a legacy business to control how the market will evolve. You could call it

18

distributed creativity, but that would make it sound as if the network was producing the

creativity. It‟s the other way around. End-to-end makes it possible to tap into the

creativity that is already distributed everywhere.

The legacy of such technical self-regulation is that minimal direct government

interference has been seen. [75]

The self-regulatory bodies are international in character

and were begun as non-commercial self-regulatory organizations (SROs). [76]

The end-

to-end principle has dictated that any content control be embedded in code by the

content creator, and filtered by browser software installed and controlled by the end-

user.

If innovation is typically both user-distributed and user-driven, the implications are that

innovation is encouraged by interoperability and open access: in general, ensuring that

content can be freely shared between those users. This view is in some conflict with

content and network owners‟ desire to be recompensed for provision of local loop

upgrades and has led to this animated debate. Note that content providers pay for their

traffic to be carried by backbone ISPs, on a best effort basis, and the argument is about

ISPs wishing to increase those payments as a result of either enhancing or blocking

service, on a mandatory or opt-in basis (clearly a mandatory blocking service for those

refusing to pay an extra toll is the most capricious of these possibilities, as we will see in

the „Madison River‟ case). Lemley and Lessig claim that innovation at the edge of the

network is opposed by traditional media and network businesses, as it makes business

cases based on controlling distribution bottlenecks redundant: where there is peer

sharing, there is less opportunity for traditional bottlenecks and therefore control of

revenues. However, the inverse applies also: without some means to secure revenues

for the increased bandwidth necessary for Web 2.0 type applications to flourish, do

network operators have an incentive to upgrade? Ed Whitacre of AT&T famously made

this claim. [77]

Of course that also can lead to a type of „arms race‟ as P2P networks

encrypt all traffic to prevent inspection, in the same way that firewalls on Intranets were

evaded using Port:80 and other techniques. [78]

Odlyzko and Levinson refute many of the arguments for fine-scaled charging which

underlie the architecture of IP Multimedia Subsystem (IMS) and QoS. They note that:

Technology appears to be making fine-scale charging (as in tolls on roads that depend

on time of day or even on current and anticipated levels of congestion) increasingly

feasible. Standard economic theory supports such measures, and technology is being

developed and deployed to implement them. But their spread is not very rapid, and

prospects for the future are uncertain … the case for fine-scale charging is not

unambiguous, and in many cases may be inappropriate. [79]

19

An emphasis on users as citizens and participants challenges conventional economic

views of the debate – where rights are tradable, the emphasis on competition wins out

in a neo-classical worldview, though with acceptance that information can improve

markets (but a bias in favour of self-regulatory means of achieving more perfect

information). For instance, it is conventional sound analysis [80]

that „The lack of a market

failure in the wireless industry suggests that such regulation would be completely

unwarranted. Consumers consistently benefit from increasingly lower prices and more

features.‟ This analysis also extends to the wireline industry, where there is competitive

entry of a type, at least in business markets and for European consumers, and lower

prices for higher speeds. A continual controversy in telecoms concerns the causation for

these benefits: does the acceleration of benefits from technology hide an oligopolistic

and uncompetitive industry structure? If that is so, and the majority of benefits flow over

from more competitive industries and more innovative platforms – notably computing

and the Internet – then should telecoms move to a computing or Internet type regulatory

structure, with emphasis on prior open standards and ex post competition regulation? [81]

This argument was used by Lemley and McGowan in the late 1990s, and has been

applied to telecoms over a lengthy period, from the commons perspective by Benkler,

Lessig and others, and recently from a more market-oriented approach by Werbach and

Weiser. Lemley and Lessig state: „The FCC‟s presumption should be against approving

mergers or policies that threaten these design principles, without a clear showing that

the threat would not undermine the Internet‟s innovation.‟ [82]

It was claimed by some pioneers such as Perry Barlow that the Internet was a global

phenomenon beyond nation-state control. It was suggested that the „Internet‟ is an

unbounded cyberspace in which borderless any-to-any communications is possible. In

view of its origins, this appears a curious statement. The Internet is a creation of

governments and educational institutions, and continues to be regulated by

government, and indeed by private corporations as a proxy for government. It can be

whatever users, governments and corporations decide it should be. There is no inherent

bias towards openness or closure of the network that cannot be undone by critical

decisions at this juncture. I note the opinion of Tim Wu, who coined the term „network

neutrality‟, and his bias towards markets in his work: „It is absolutely not a call for

comprehensive regulation or nationalization of the wireless industry. The perspective is

that regulation, if necessary, should be a last resort.‟ I also note the analysis of the head

of the Canadian regulator, the Canadian Radio-television and Telecommunications

Commission (CRTC), Konrad von Finckenstein, who states: „Fundamental issues of

technology, economics, competition, access and freedom of speech are all involved.‟

Consider an excellent summary of these views offered by de Beer: [83]

20

Too much of the network neutrality debate seems to be cast in economic terms, with

advocates on either side trying to establish whether openness or deregulation will be

most effective for competition and innovation … My intention is not to detract from the

economic arguments, but to bring in culture as another lens through which to view the

network neutrality issue.

Commissioner Reding stated on 6 May 2009 in the European Parliament prior to its vote

on the new telecoms law: [84]

Even though traffic management can allow premium high quality services to develop

and can help ensure secure communications, the same techniques may also be used to

degrade the quality of communications or other services to unacceptably low levels.

That is why, under the new EU rules, national telecom authorities will have the power to

set a minimum quality level for network transmission services so as to promote net

neutrality and net freedoms for European citizens.

So far, so unremarkable. Note that it is up to the national regulators whether they want

to take on this extra task, which is technical, fiddly and may not be a high priority for

some – or indeed all – of them. She then goes on to paint that policy in political rather

than technical terms:

The fourth element I would like to underline is the recognition of the right to Internet

access. The new rules recognize explicitly that Internet access is a fundamental right

such as the freedom of expression and the freedom to access information. The rules

therefore provide that any measures taken regarding access to, or use of, services and

applications must respect the fundamental rights and freedoms of natural persons,

including the right to privacy, freedom of expression and access to information and

education as well as due process.

She is stating an extension of a principle for European telecoms regulation: the existing

rules for common carriers – the phone monopolies – that require them to connect all

users on demand to the phone network and allow them to communicate without prior

censorship on their speech, will be extended to require carriers to permit Internet users

to maintain their access to the open Internet, subject to the law as enforced by courts.

The open Internet access right – in reality based on the payment by citizens to those

carriers – maintains the principle that the telecoms network is a common carrier. What

is interesting in this debate is not simply that technology lets carriers fiddle with citizens‟

communications, but what motives they have for doing so, and how robust is the

regulatory response to any ill-considered tampering.

21

Structure of the Book

Who’s to Blame?

My starting point in unpacking the debate is that network neutrality has a positive or

forward-facing element related to ISPs investing in faster Internet access and a negative

or backward-facing element related to blocking or throttling on existing networks. I

extend this negative net neutrality analysis as a proposal for „net neutrality lite‟ for

Europe, which the United States and other countries may also find useful. It is a

pretence to claim that different countries can be independent of each other in these

policies, when the content and router worlds are particularly globalized – what Disney,

Google and Cisco do in the United States must necessarily have a massive impact on

Europe, in particular because most of the credible market entry competition to European

incumbent behemoths comes from the United States. [85]

As this introduction begins to uncover, „net neutrality‟ is a deceptively simple phrase

hiding a multitude of meanings. First, it has to be unpacked to discover that it comprises

two separate non-discrimination commitments. Backward-looking „net neutrality lite‟

claims that Internet users should not be disadvantaged due to opaque and invidious

practices by their current ISP. Forward-looking „positive net neutrality‟ describes a

practice whereby higher QoS for higher prices should be offered on fair reasonable and

non-discriminatory terms to all-comers, a modern equivalent of common carriage. It is a

more debatable principle, with many content providers and carriers preferring exclusive

arrangements. That is what the first three chapters seek to establish. In Chapter 1, I

introduce broadband supply and investment and blocking technologies and

requirements, in Chapter 2 the types of „negative‟ discrimination that may occur and the

development of the debate in both the United States and Europe, and in Chapter 3

issues of QoS, user-generated and/or distributed content. In Chapter 3, I consider

„positive‟ net neutrality in detail.

I argue that the entire net neutrality debate has been set up as a false dialogue of the

deaf between the net neutrality absolutists on one side and the net neutrality refuseniks

on the other. The absolutists are caricatured as favouring absolutely no traffic

management, or prioritization, thus leaving the Internet in a type of primordial Garden of

Eden in which even content delivery networks such as Akamai that push content closer

to end-points (which we explore in Chapter 2) are seen as suspiciously „efficient‟ in

delivering packets closer to the end-user. Not surprisingly, Lessig and Lemley‟s

argument is portrayed by net neutrality opponents as nostalgia for a halcyon „Golden

Age That Never Was‟. Consider: Sandvig states that there has always been traffic

management of various sorts on the Internet; Crowcroft robustly establishes the network

22

engineers‟ perspective that – far from net neutrality being an end-goal – QoS on the

Internet has always been a kind of Holy Grail to prioritize and more robustly deliver

higher-value services, from the network operator‟s point of view; Clark, one of the three

authors of the original „end-to-end‟ principle, states that he would have more carefully

chosen his terms had he realized the „Talmudic‟ reverence with which his engineering

argument was viewed by social scientists. [86]

Traffic management on the Internet is a

fact of life, and to suggest otherwise can clearly invite – and has created – a storm of

derision for those of the extreme net neutrality persuasion.

The refuseniks have their man of straw to deride. However, if it is true that there has

always been some kind of QoS on the Internet, it is equally true that for perhaps twenty-

five years, carrier-class QoS was to be introduced „next year‟. Of course private

networks and virtual private networks have dedicated circuits, and often dedicated

networks, devoted to private clients of one type or another. Public networks have not

achieved truly differentiated QoS to date, and perhaps never will. So there is another

man of straw: the continued attempts to create QoS on a „best efforts‟ network could be

a hopeless quest, as searches for Holy Grails often are. If that is true, then increasing

capacity many fold, as has been achieved by the super-fast Internet2 for university

networks, would be the most useful resource in which to invest.

This leads to a further speculation. If the quest for the Holy Grail of QoS truly is doomed

to fail, then the net neutrality absolutists are actually tilting at windmills themselves, and

any move away from net neutrality is a waste of money. In the fullness of time, and

assuming a belief in the self-correcting folly of markets even in the dark days of 2009,

ISPs will see the error of their ways and return to the light of an unfettered open

Internet. Two obvious problems present themselves: first a great deal of time, effort and

money will have been wasted in the meantime with irreparable harm (as so many harms

are, though far from fatal) to innovation and end-user remixing; second, given the sunk

costs in investing in DPI equipment and its management, will ISPs have become liable

for their content as publisher not carrier by that point? The entire liability protection

regime established in the late 1990s might have unravelled by then. Politicians must

make these decisions, but how can we expect them to understand the issues, or to

arrive at sensible solutions? The entire field is confused, and such principles as they

exist are only at the extremes. I will examine this in Chapter 1 and Chapter 4 and 5.

I have to mention copyright if only to indicate a further caricature within the debate. It

has two elements. The first is that all high-capacity consumer users of the Internet are

file-sharing. The second is that this file-sharing is illegal, as it is sharing of copied

„pirated‟ files that breach others‟ copyrights. Both are of course highly contentious, but

as with all half-truths, the kernel of truth does lie inside that there is large-scale „theft‟ of

23

„pirated‟ material, to use the terms preferred by the copyright industry. I write this as The

Pirate Bay, a rather blatant example of a site offering access to copyrighted material for

free download, thus breaching copyright law, has seen its principal officers both fined

and sentenced to prison by a Swedish court of first instance, though with appeal based

on the judges‟ alleged professional bias and connections to the copyright industry. Ever

since the prosecution of Napster and its founder Shawn Fanning in 1999, the copyright

industries have waged a war on copyright violators online. This decade-long battle

against „pirates‟ (which conjures images of Somali teenagers with machine guns rather

than Western college students in dorm rooms [87]

) has of course proved a fairly fruitless

exercise for copyright holders, and a burdensome and expensive exercise for ISPs.

End-users have typically had their rights broached in some form not least in breach of

privacy, and in communications that expose their activities within their households

(identity within families is always an early casualty of Internet access disputes). The

solution offered is to chase individual infractors and to further attack a class of data

traffic, P2P networks, or within that, a sub-class, BitTorrent. This can be achieved by

various blunt tools, including throttling, torrent seeding, disconnection at fifteen-minute

intervals, and so on. In France, the proposal to establish an administrative solution to

close down the connections of persistent violators (so judged by ISP and copyright

industries rather than courts) has been defeated in April and June 2009 (the so-called

HADOPI laws), while controversy continues over European Parliament amendments to

consumer protection legislation online. We will return to this issue in Chapters 4–6. In

Chapter 4, I broaden the debate to consider the types of filtering and blocking that ISPs

have traditionally engaged in, and the legal liabilities that they have incurred. I consider

the regulatory challenges in dealing with „lite‟ net neutrality issues in Chapters 5 and 5

with the specific European and UK responses to content regulation, demonstrating

creeping control and a reversal of the presumption that ISPs do not carry liability for

their content. My discussion about how net neutrality regulation in the United Kingdom

fits into the reviewed 2009 Electronic Communications Framework (the Directives which

direct European NRAs to regulate ISPs) may be helpful. The book‟s argument is

illustrated by three reviews of existing policies: an extended discussion of British policy

in Chapters 6, a discussion of EC policy in Section 5 and an early discussion of US

policy in Chapters 2 and 3, because the US debate has cast its light and darkness on

the later debates elsewhere.

I have some partial solutions but no panacea. Mobile Internet claims the same special

protections from regulation that its forerunner, mobile termination, claimed, to enable

walled gardens to flourish. In Chapter 7, I analyse the problems inherent in regulating

the mobile Internet for net neutrality, with emphasis on pre-existing content and price

discrimination and regulation in Europe. Finally in Chapter 8, I consider the potential for

24

this debate to develop towards a form of net neutrality „lite‟. Transparency and

investigation are easy wins, principles that regulators can only discount on grounds of

ignorance („net neutrality is a solution in search of a problem‟) or resource depletion

(„we don‟t regulate the Internet, it is too complicated‟). It improves competitive forces,

such as they are, so the pro-market advocates surely have little to disagree with,

especially as it is to be achieved through co-regulatory means at lowest cost to ISPs.

Co-regulation is a prevalent but awkward compromise between state and private

regulation, with constitutionally uncertain protection for end-users and a worryingly large

latitude for private censorship, which has been increasing throughout the last decade

even as the law declares ISPs to be „three wise monkeys‟, as discussed in Chapter 4.

Any solution needs to be holistic, considering ISPs‟ roles in the round, including their

legal liabilities for content filtering. I will grapple with the dilemma that I may be giving

the ISPs a free lunch: the appearance of a solution without even a partial remedy for

end-users. I create trouble for myself by admitting to both economic and normative

rationales, each pulling in different directions, and attempt unsuccessfully to square that

circle. That is why I only recommend net neutrality „lite‟ – the easy backward-looking

solution.

This is a policy area with no right answers that offer perfect solutions. Of course the

Internet should be open to all, but of course private investment is the critical component

in building a faster Internet. Of course universal service should be supported, but of

course there must be some minimum access to the open Internet for all, whether they

use a mobile 3G connection or a fast Internet Protocol Television (IPTV)-enabled

premium service. If it says „Internet service‟, it should do what it says on the tin, offer an

open Internet (alongside walled gardens if expressly advertised as such). I am happier

limiting my solution to emphasize the complexity of the problem than trying to claim a

one-size-fits-all solution. Net neutrality is an issue with potentially profound

consequences, and cannot be entirely left to market actors, however neutral or benign

their motives.

Notes

[1] MP3 from 8 June 2006. Note the 44th President is not opposed to QoS, just to differential

contracts for that QoS. He continues: „Allowing the Bells and cable companies to act as gatekeepers with control over Internet access would make the Internet like cable. A producer- driven market with barriers to entry for website creators and preferential treatment for specific sites based not on merit, the number of hits, but on relationships with the corporate gatekeeper. If there were four or more competitive providers of broadband service to every home, then cable and telephone companies would not be able to create a bidding war for access to the high- speed lanes. But here‟s the problem. More than 99 percent of households get their broadband services from either cable or a telephone company.‟

25

He went on: „There is widespread support among consumer groups, leading academics and the most innovative Internet companies, including Google and Yahoo, in favor of net neutrality. And part of the reason for that is companies like Google and Yahoo might never have gotten started had they not been in a position to easily access the Internet and do so on the same terms as the big corporate companies that were interested in making money on the Internet…‟.

[2] The Internet is a „networks of networks‟ that connects users by sending packets of bits (digital

data) from any point on that network of networks to any other point. Kahn and Serf adopt the broad Federal Networking Council definition of the Internet: „Of particular note is that it defines the Internet as a global information system, and included in the definition, is not only the underlying communications technology, but also higher-level protocols and end-user applications, the associated data structures and the means by which the information may be processed, manifested, or otherwise used.‟ It is defined by its inventors, Cerf et al. at http://www.isoc.org/Internet/history/brief.shtml. See further Kahn and Cerf (1999).

From the October 24, 1995, Resolution of the US Federal Networking Council in Kahn and Cerf, op cit: „The Federal Networking Council (FNC) agrees that the following language reflects our definition of the term “Internet”. “Internet” refers to the global information system that – (i) is logically linked together by a globally unique address space based on the IP (IP) or its subsequent extensions/follow-ons; (ii) is able to support communications using the Transmission Control Protocol/IP (TCP/IP) suite or its subsequent extensions/follow-ons, and/or other IP-compatible protocols; and (iii) provides, uses or makes accessible, either publicly or privately, high level services layered on the communications and related infrastructure described herein.‟

For most domestic, consumer users, it is a combination of the World Wide Web, a graphical interface that permits hyperlink surfing („clicking‟ from one webpage to another) and electronic mail. However, it is actually the most prevalent and fastest growing information sharing network ever devised. It grew from a small scientific network into a network accessed by almost a billion people in less than 10 years (1994–2003), only exceeded in rate of diffusion by mobile telephones. Numbers of Internet users have consistently been under-estimated in official statistics, and there is no accurate estimate for the global total. There are no definitive totals of Internet users worldwide, unsurprising given the numbers of email accounts, use of cybercafes, numbers online through a third party subscription (e.g. work, school, library). See Global Internet Statistics http://www.glreach.com/globstats/index.php3 and GSM Association http://www.gsmworld.com/news/statistics/index.shtml. However, it is meaningless to give specific numbers for a book – there are well over a billion Internet users by mid-2009 and well over 2 billion mobile phone users.

[3] I am Senior Lecturer at the University of Essex Law School (since 2008), and formerly

Lecturer (2007–8) and Fellow (2005–7), having also been a Research Manager/Associate at Oxford‟s Internet Institute and Programme in Comparative Media Law and Policy (2003–5), and previously Lecturer in European Law at the University of Warwick (1997–2000).

[4] I consulted for the Chief Executive of the Independent Television Commission, a forerunner of

UK regulator Ofcom, in 2000, lobbied Oftel, another forerunner in 2001–2, as Director of Regulatory and Government Policy for MCI WorldCom UK Ltd (yes, WorldCom), was General Counsel of Shortmedia Ltd, a video-on-demand start-up in 2000–1, consultant for telecoms strategy boutique Re:Think! In 2001, and for RAND Corporation‟s Cambridge and Brussels

26

offices, I worked for Ofcom as well as British Telecom and content providers such as Google Inc. in 2005–7. I have seen UK communications policy from most sides.

[5] I have consulted for the European Commission as well as the Netherlands and Irish

governments in 2005–7, and for the Council of Europe in 1999.

[6] I was Research Fellow at Harvard‟s Information Infrastructure Project (1999–2000),

participated in the Harvard/Swiss Re conferences (2002–7), and Wharton Media Law Colloquiums (2006–8) and have given many papers since 1997 at the Telecoms Policy Research Conference, serving on the Programme Committee for the 2007 and 2008 conferences. I was also a very active member of the Cambridge-MIT Institute Communications Research Network 2004–7. I also have long-standing links with both USC and Columbia University researchers in the field.

[7] I am a Research Fellow at both GLOCOM and Keio Universities and a frequent visitor to

Japan, including several visits sponsored by the British Academy, Japanese Society for the Promotion of Science and Keidanren 21st Century Institute.

[8] Berners Lee (2006) at http://dig.csail.mit.edu/breadcrumbs/node/144.

[9] The comparison is carefully drawn, as congestion modelling for IP networks and roads has

been carried out with great policy effect by Kelly (2003).

[10] Still the only option for the vast majority to the developing world, where mobile subscribers

and WiFi users outnumber fixed broadband subscribers exponentially – for instance by at least 20 to 1 in India.

[11] There are 8 bits in a byte.

[12] See Lessig (1999b); Lemley and McGowan (1998); Lemley (1999, 2000) in Marsden (ed.).

[13] Warwick University had only a theoretical capacity to deliver real-time video conferencing all

those years ago. The first article I give to my LL.M. introduction class is Samuelson (2000), asking them which of the five challenges remains most relevant. (It‟s a trick, they all do!).

[14] Lemley and Lessig (1999) at Paragraph 4.

[15] Lemley and Lessig (1999) at Paragraph 22. Though note that end-to-end has prevented

many Internet architectural innovations including QoS. Dave Clark, one of the originators of the end-to-end principle, has pleaded for less zealotry on the part of lawyers interpreting his work as though it were a „law‟: see Clark and Blumenthal (2001) in a very thoughtful restatement of the ideas.

Clark more recently expressed his views that end-to-end anonymity was actually a bad mistake, particularly in the spam context. See: http://www.cambridge-mit.org/cgi- bin/default.pl?SID=5&CALEVID=192.

[16] For a more free-market empirical observation, see Hazlett and Bittlingmayer (2001), Marsden

(2000b) and most recently Hazlett and Caliskan (2008).

27

[17] Lessig (1999a).

[18] Marsden (2004a: 5).

[19] The great man did concern himself with both vertical integration and moreover the UK public

service broadcaster, in his LSE days. See Coase (1937, 1950) at http://coase.org/coasepublications.htm.

[20] To understand the nature of the relationship between regulation and competition law in the

late 1990s, see Sauter (1997); Larouche (2000); Marsden ed. (2000) (especially chapters by Cave, Collins and Barnes); Cowie and Marsden (1999).

[21] Generally, see Walden (2009). Early case law is covered in Ungerer (1996a); Jauk (1999).

[22] On US policy generally, see Lichtman, Shelanski and Weiser (2006) at

www.law.duke.edu/fac/benjamin/telecom/.

[23] IP/99/413 (1999).

[24] See IP/00/373 (2000), resolving Case No COMP/M.1795.

[25] This issue arose in Europe in three mergers: AOL and Time Warner, WorldCom and MCI,

and WorldCom and Sprint. See Case No COMP/ M. 1845 AOL/Time-Warner; IP/98/213 (1998); IP/98/639 (1998); Case IV/M.1069 - WorldCom/MCI 99/287/EC; Case COMP/M.1741; MCI WorldCom/Sprint, D.Comm. June 28, 2000, 2003 OJL 300/1; Case T-310/00 MCI, Inc. v. Commission.

[26] Directive 96/19/EEC.

[27] See LeGates (2000); Hurley and Keller (eds.) (1999).

[28] See the case law: Sealink v. B&I Holyhead: Interim Measures [1992] 5 CMLR 255; Sea

Containers v. Stena Sealink Commission Decision 94/19 [1994] OJ No L 15/8; Commission Decision 89/205/EEC in Case IV/31.851 - Magill TV Guide/ITP, BBC and RTEOJ L 78, 21.3.1989, p. 43; C-241-242/91; Magill v. RTE and Commission (1995)4 CMLR 718 C-7/97; C- 241 / 91P & C-242/91P, Radio Telefis Eireann v. Commission („Magill‟), [1995] ECR, I-743; Oscar Bronner GmbH&Co. KG v. Mediaprint Zeitungs- und Zeitschriftenverlag GmbH&Co KG [1999] 4 CMLR 112; C-418/01 IMS Health v. NDC Health 29th of April 2004, ECR 2004. Analysis provided in Cowie and Marsden (1999); Treacy (1998); Bergman (2000); Schmidt (2002).

[29] On economic issues in telecoms generally, see Laffont and Tirole (2001); Majumdar,

Vogelsang, Cave eds. (Vol. 1 2002; Vol. 2 2005); Buiges and Rey (2004); Cave (2004).

[30] C(2000)1059.

[31] On the chequered progress of NRAs in implementing the 1998 and 2002 frameworks, see

British Institute of International and Comparative Law (2004); Geradin and O‟Donoghue (2005); Andersen (2005).

28

[32] United States v. Microsoft Corporation (2000); in re United States of America v. Microsoft

Corporation, Civil Action No. 98-1233 (CKK).

[33] COMP/C-3/37.792 Microsoft (24.03.2004); C(2004)900.

[34] On Trans-Atlantic differences, see Naftel and Spiwak (2001); Marcus (2005).

[35] Ungerer (2000: 24).

[36] IP/98/707 (1998); IP/99/298 (1999).

[37] See Ungerer (2000) at footnote 93: See also[the] recent FCC (Federal Communications

Commission) decision on the AT&T/MediaOne merger where the FCC insisted on divestitures, in order to decrease the effect of the merger on the cable TV market, and noted that it expected „AT&T to fulfil its voluntary commitments to give unaffiliated ISPs (ISPs) access to its cable systems to provide broadband services to consumers‟. It also noted „that AT&T has entered a proposed consent decree with the U.S. Department of Justice, which requires the merged firm to divest its interest in the cable broadband ISP Road Runner and to obtain Justice Department approval prior to entering certain types of broadband arrangements with Time Warner and America Online‟.

[38] European Parliament legislative resolution of 6 May 2009: 16498/1/2008 – C6-0067/2009 –

2007/0249(COD).

[39] SPEECH/08/561.

[40] Reding continued SPEECH/08/561: „Professor Martin Cave has recently calculated that the

present lack of a single telecoms market comes at a very high price for Europe‟s economy. According to him, the additional cost of regulatory fragmentation in telecoms is €20 billion per year for Europe‟s businesses. Commission experts believe this figure to be still a very conservative estimate … But it is clear that at present, it remains difficult for, say, a French operator to invest in Spain if regulatory decisions on next generation access differ substantially in both countries. And I compare with interest your comments on regulatory issues with that of your subsidiaries in those Member States where they are in the position of a new market entrant … This [type of regulatory inconsistency] is exactly the kind of problematic situation Professor Cave refers to when he talks about the cost of poor and inconsistent regulation.‟

[41] Lessig‟s further work may be more well known, but his former student, the now masterful Tim

Wu also has an excellent record of scholarship investigating control of content and freedom of expression on the Internet in the United States, Europe and China. See Goldsmith and Wu (2006).

[42] Sandvig (2007).

[43] It is often forgotten that Enron‟s troubles were highlighted by the Enron Broadband non-

business whose revenues were wildly overstated in 2000. After WorldCom‟s $11b fraud was later uncovered, in June 2002, global stock markets fell 25%, as Arthur Andersen collapsed and many of the world‟s largest companies had to restate earnings. It was a forewarning of the 2008 implosion of banks. Not only those companies and Global Crossing, but also competitors such as Level3 went into Chapter 11 bankruptcy protection, from where they aggressively – almost

29

suicidaly – cut prices to customers, thus playing „beggar thy neighbour‟ tactics against their competitors such as the UK‟s Cable & Wireless and Energis. For the US fall-out see Goldstein (2005), also see http://telefrieden.blogspot.com/2009/02/non-lesson-from-telecom-frauds.html for a comparison with the present day.

[44] I exclude the United Kingdom, as its cable provider was deep in financial trouble, and the

pace of broadband rollout was funereal.

[45] Cairncross (1997), stating that policy must adapt to the loss of local voice telephone

monopolies, while dealing with the „Five P‟s‟ of the Internet: policing, pornography, privacy, protection and property. She also feared the monopolistic tendencies of the industry, as revealed by the anti-trust charges against Microsoft: „Competition clearly does not come naturally in communications.‟

[46] Even if many consumers only wanted their wireless phone and broadband, what is termed

„naked DSL‟ which rival ISPs can sell on to customers without the twentieth century phone line paraphernalia. As I write, this is available in France and other EU countries, but not the United Kingdom and the majority of the 27 Member States.

[47] Riley and Scott (2009: 3).

[48] Recording Industry Association of America v. Verizon Internet Services 2003 U.S. Dist. Lexis

681, 240 F. Supp. 2d 24 (D.D.C., Jan. 21, 2003), reversed, 351 F.3d 1229, Case No. 03-7015 (D.C. Cir., December 19, 2003) cert denied 125 S.Ct. 309 (2004).

[49] See Croxford and Marsden (2001) for contemporary details.

[50] Wu (2003).

[51] Hulme-Jones (2009).

[52] Ridley and Scott (2009: 3) add: „The first DPI devices were used for manual troubleshooting

of network problems and to block viruses, worms and Denial of Service attacks. Initially, DPI was not powerful enough to monitor users‟ Internet communications in real time. But today, DPI is capable of far more than security – it enables new revenue-generating capabilities through discrimination. Historically, Internet communications were processed using only information in the header, because only that information is needed to transfer packets from their source to their destination. By contrast, DPI technology opens and reads the data field in real time, allowing network operators to identify and control, at a precise level, everyday uses of the Internet…‟

[53] See Marsden et al. (2006) at Appendix A: A Simple Game for ISP Choice, modelled by

Jonathan Cave – showing that early adopters with a propensity to act as servers for P2P networks are exactly the kind of customer that no ISP wants, and that therefore all ISPs logically choose to throttle such uses.

[54] Crowcroft (2007).

30

[55] Public remarks of discussion between UK and French regulators at ENST conference 29 May

2007 in Paris.

[56] With walls sufficiently high that one might term them „Forbidden Cities‟. The reference is to

the Chinese Emperor‟s official residence in Peking until 1924, and I acknowledge fully the analogy to the „Great Firewall of China‟ – on which see Zittrain and Edelman (2003) and Goldsmith and Wu (2006).

[57] Van Schewick (2005).

[58] Kocsis and De Bijl (2006) have proposed a game theoretical perspective to analyze such

incentives similar to Appendix A in Kocsis and De Bijl (2006).

[59] P2P networks carry malware, spyware, spam and other unsolicited and potentially harmful

content, but so does SMTP email, in fact the latter is generally considered a more ubiquitous danger.

[60] See Greenberg and Veytsel (2006) and Greenstein (2006).

[61] Examples abound, for recent examples see http://chrismarsden.blogspot.com/.

[62] See Clayton (2005); Brown (2008); Pfleeger and Pfleeger (2006).

[63] As there is usually only a telephone line, and sometimes a cable line (depending on cable

industry development), into each domestic household, there are only two possible competitors unless one or both lines are shared with other rivals.

[64] Williams (2007).

[65] Meyer (2007).

[66] A common ITU term for all-IP networks, which are replacing the current telephony networks.

[67] Frieden cites Code Monkey (2006): „What the ISPs don‟t tell the public is that there are no

free-riders among the content companies. They pay handsomely for their bandwidth. In fact, they are the true bread and butter for the major telecoms and ISPs. The reason that this “Network Neutrality” controversy exists today is that ISPs don‟t want to admit that their whole business model is flawed.‟

[68] Davies and Banks (eds.) (2003).

[69] See http://arstechnica.com/articles/culture/Deep-packet-inspection-meets-net-neutrality.ars/2.

[70] See Frydman, Hennebel and Lewkowicz (2008) at http://ssrn.com/abstract=1282826. Also

note the work of the OpenNetInitiative, the Chilling Effects clearing house and other academic- NGO initiatives to track censorship on the Internet: http://opennet.net/ http://www.chillingeffects.org/ http://www.eff.org/ and in Europe organizations affiliated to the European Digital Rights Initiative: http://www.edri.org/about/members

31

[71] Mayer-Schonberger (2008). Also more generally on law and ecomomics‟ place, see Mackaay

(2006).

[72] Berners Lee, with Fischetti (1999).

[73] Lessig (1999d).

[74] Berners Lee, with Fischetti (1999).

[75] Gillett and Kapor (1996) in Kahin and Keller (eds).

[76] Lemley (1999) and Lessig (1999c).

[77] Business Week International Online Extra (2005).

[78] See generally Pfleeger and Pfleeger (2006).

[79] Odlyzko and Levinson (2007: 3).

[80] Wallsten (2007) at SSRN: http://ssrn.com/abstract=976749

[81] This debate was current even a decade ago at the beginning of the open access debate: see

Woroch (2002); Weiser (2000); Speta (2000).

[82] Lemley and Lessig (1999: 1).

[83] De Beer, Jeremy (2009: 24.3) states: „Both supporters and opponents of regulated network

neutrality have discussed the principle as a First Amendment issue for its connection to freedom of online expression (Yemini 2008, May 2007). Its egalitarian and participatory underpinnings have even been connected to fundamental theories of distributive justice (Schejter and Yemini 2007).‟ De Beer‟s work alongside that of his colleague Michael Geist at the University of Ottawa is doing much to shape the arguments around network neutrality as both an economic and a human rights issue. De Beer, Jeremy (2009: 24.3) states: „Both supporters and opponents of regulated network neutrality have discussed the principle as a First Amendment issue for its connection to freedom of online expression (Yemini 2008, May 2007). Its egalitarian and participatory underpinnings have even been connected to fundamental theories of distributive justice (Schejter and Yemini 2007).‟

[84] Reding (2009 undated).

[85] It is reminiscent of European attempts to resist the Netscape and Explorer browsers‟ decision

to set a default of accepting cookies for e-commerce in the 1990s.

[86] See Sandvig (2007); Crowcroft (2006); Clark and Blumenthal (2007).

[87] See Palfrey and Gasser (2008); Tapscott (1999, 2008). On the specifics of file-sharing, see

Danay (2005).