Google reminds Verizon of net neutrality differences

Of the Internet and Artificial Terrariums

Rick Whitt, Google, Feb. 23, 2010

When I was in second grade, our elementary school sponsored a science fair.  One of my best friends brought in a large glass jar, filled with mud and leaves and a few bored-looking insects, which she proclaimed a "terrarium," a small-scale, self-enclosed natural environment.  Of course, when she forgot to include some water and punch some air holes in the lid, the terrarium quickly turned into a miniature mausoleum.

I was reminded of that glass jar as our broadband provider friends have begun to take up the meme of "the Internet as an ecosystem."  Verizon in particular has been employing that concept in public filings at the FCC, and in several recent blog postings.  These include relatively benign statements like "All players, big and small, play vital roles," "All parts are interrelated in ways not always obvious, but always important," and "Value the trees, but see the forest," which sounds more like a Zen saying than a policy view.

However, Verizon's folks have made a few other observations:
(1) -- One should not try to classify and treat differently the entities in an ecosystem, as "distinctions among these categories rapidly erode and lose their meaning." The FCC should not create "artificial regulatory silos" of "devices," "applications," "content," and "networks." In fact, "attempting to treat some parts different than others is a recipe for disaster for the whole." (VZ 2/1/4/10 blog post, Verizon 1/14/10 NN comments at 1-3)
(2) -- "The Internet ecosystem is very robust and self preserving...." (VZ 1/14/10 blog post)


On the surface, the ecosystem analogy has some appeal, and does generally track some cutting-edge theories about economic systems as modeled on biological systems.  However, there are some important ways that the analogy breaks down in at least two of the lessons drawn by the folks at Verizon, and at the same time helps explain our fundamental views -- and differences -- in the open Internet debate.

First, treating some parts of an ecosystem different from other parts is not a "recipe for disaster;" it is just plain common sense.  One of the main characteristics of an ecosystem is its rich diversity; from flora to fauna, carnivores to vegetarians.  In the context of the Internet, there is also admirable diversity between network providers and content providers -- those who employ atoms and those who employ bits -- as embodied in corporations of all shapes and sizes, individual users, non-profit groups, and of course the government.

However, while each entity occupies a niche in the ecosystem, no two niches are the same. 

And not all players play vital roles; some are more vital than others, operating as "keystone" species upon which many others depend. In fact, broadband serves a prime "keystone" role in the ecosystem: it is an essential input to all other "species" of communications, information, and entertainment sources; it is a scarce resource in an otherwise robust environment; and it serves a unique control function no other species can match.  Staying with the metaphor, one can think of broadband as the soil or the water that nourishes the rest of the Internet "ecosystem."

Analogies and metaphors can only take us so far.  To better understand the Internet, I would prefer using a conceptual model that actually tracks to the realities of technology and market structures -- namely, the layered nature of the Internet.  Under this model, entities are classified based on what functions they perform in the network, whether at the physical layer, the ISP layer, the applications layer, or the content layer.  Simply put, you are what you do.  Thus, for example, even though the physical layers are vital to the activity of those in the upper applications and content layers, that certainly does not amount to an equivalence, in terms of the respective roles played in the market.  Instead, the physical layers support all the activities that "ride on top," including the world of the Internet.

Obviously a good ecologist would view and treat different parts of an ecosystem in disparate ways, depending on those very functional distinctions. Similarly, just because Verizon also provides online applications does not change the nature of its networking business.  And just because Google intends to pursue a broadband testbed does not mean it should no longer be seen also as a software applications company.  It is just a fact that broadband networks are not wireless app stores, or search engines, or an online social community.  Each function needs to be understood in terms of the larger whole, but the whole does not consequently swallow up salient distinctions between the parts. 

Second, I take issue with the suggestion that economic systems are inherently self-regulating, with little to no need for any "outside" influence.  This sounds like a modern version of Adam SmithAdam SmithPentagon urges Congress to move 6M for missile defense Pass the Protecting Data at the Border Act Overnight Defense: Mattis makes surprise visit to Afghanistan | Army general deploys to Puerto Rico to oversee hurricane relief | Senate panel advances Joint Chiefs chair's nomination MORE's famous "invisible hand" of the market.  As the global financial disasters of the last several years have amply demonstrated, however, the more extreme version of this image simply is not true.  No economic system is completely self-regulating.  We need human-made institutions -- laws, regulations, standards, norms -- to make economic systems function properly.  Without the fundamental trust and equity brought about by property laws, contract rights, labor laws, and consumer safeguards, the "free market" cannot hope to survive.

The Internet is no different.  While still a marvel of self-regulation on many fronts, obviously the Internet is not a government regulation-free zone.  Activities that are unlawful in the offline world are just as unlawful online.  And consumers still need to be protected in their daily commercial transactions, whether someone steals their credit card at a grocery store or misappropriates it during an online sale.  The right question to ask is not whether any governing institutions are necessary, but what kind -- government, industry, or something else -- and in what way.

We believe as a general matter that the Internet of bits (software-generated applications and content and services) is more robustly competitive, and open to dynamic future competition, than the broadband infrastructure of atoms (the physical on-ramps taking consumers to and from the Internet). 

As a result, the latter market requires more concerted government attention than the former.  Nonetheless, the Federal Trade Commission has assumed the role of "cop on the beat" in the online commerce world, in areas like online advertising and privacy, and it is a role that we accept.  To Verizon's credit, they recognize the need for appropriate "safeguards" against bad acts by broadband providers that harm users.  However, at the same time, the broadband providers resist the FCC's authority, whether under Title I or Title II or anything else, to oversee the bad acts and bad actors in their sector.

In the open Internet proceeding, Google has taken the position that the federal government needs to establish tailored "rules of the road" to govern certain market practices by broadband companies.  Verizon and other broadband providers do not agree, and that is a worthy debate to have.  But it behooves all of us to have that debate on a record of empirical facts and economic analysis, and not premised on an otherwise useful metaphor stretched too thin and too far.