Why IP interconnection would break the Internet

Scott Cleland Contributor
Font Size:

As the nation continues the “IP transition” from the increasingly obsolete public switched telephone network (PSTN) to the modern broadband Internet, many are naively and unwisely recommending that common carrier interconnection regulation should replace un-regulated Internet peering agreements

The PSTN “interconnects” different entities based on compulsory, regulated, interconnection-agreements. The Internet inter-networks different entities based on voluntary, unregulated “inter-networking” or “peering” agreements. One does not “connect” to the Internet, one accesses it.

IP interconnection advocates just don’t get it. The dedicated phone network is quickly becoming obsolete. It is eventually going the way of the telegraph. Just like the telephone network made the 1800’s telegraph network obsolete because it could do so much more, the modern broadband Internet is making the telephone network obsolete because it has vastly more functionality and utility.

Those seeking to impose 1887 common-carrier, railroad regulation upon the 21st century Internet support retrogression — the antithesis of modern progress. It would move us backward not forward.

“IP interconnection” is an oxymoron. Advocates for imposing IP interconnection on the Internet via the IP transition expose a profound ignorance of Internet technology and how it really works.

First, interconnection is a telephone-technology-specific concept. It was created and designed for circuit-switching telephone technology that purposefully produces dedicated electrical circuits connecting specific telephones.

In stark contrast, inter-networking or “peering” is an Internet-technology-specific concept. It was created and designed for packet-switching Internet technology. It purposefully subdivides information into many small packets to enable much more efficient transmission. The packets get individually routed and commingled with other transmissions packets to minimize bandwidth waste. At the ultimate destination, the small packets get immediately reassembled by most any device anywhere.

Second, interconnection is all about creating and maintaining a continuous electrical circuit, while inter-networking is all about one computer network facilitating the forwarding of data packets to another computer network based on a discontinuous system of Internet protocol packet delivery.

Third, the predictable transmission path of telephone circuit-switched connections is exact opposite of the unpredictable transmission path of Internet packet delivery.

Fourth, the PSTN was designed to provide one service — voice. In stark contrast, the Internet was designed to enable any product or service one could invent. On the Internet voice is an app. Consequently, the Internet can mimic the PSTN easily and cheaply with Voice over IP (VoIP), just like the Internet can mimic broadcasting with video streaming.

Fifth, telephone technology by design is centrally-organized in a hub-and-spoke manner utilizing central office switches to connect calls. Internet technology by design is decentralized utilizing routers which simultaneously forward commingled packets between computer networks via routing tables and “best efforts.”

Sixth, since the predictable and centralized design of a telephone network naturally lends itself to predictable, 99.999% reliability and operation, it is naturally conducive to a centralized management regulatory construct like common carrier interconnection.

However, the unpredictable, “best efforts,” and decentralized Internet design is antithetical and inherently hostile to a centralized management regulatory construct like common carrier interconnection.

Seventh, interconnection is important in a circuit-switched network because one must connect at the nearest geographic point, which inherently limits competitive interconnection choices. However, inter-networking or peering with other computer networks can largely happen without respect to location so consequently an Internet entity inherently has many competitive choices to inter-network and access the Internet.

Simply, circuit-switched technology can need regulation to ensure interconnection, whereas Internet technology does not because the design of the Internet naturally facilitates voluntary and competitive inter-networking or peering access to the Internet in a location-agnostic and technology-agnostic manner.

Lastly, interconnection is an economic concept focused on accounting and billing for originating and terminating traffic. By design Internet technology does not originate or terminate connections; it breaks information into packets, commingles and forwards them for reassembly on any device anywhere.

Given the factual reality described above that interconnected telephone networks are effectively the opposite of inter-networked computer networks, trying to impose telephone interconnection rules on IP inter-networking is akin to forcing a square peg into a round hole. It predictably breaks both the peg and the hole.

So how would the FCC imposing IP interconnection requirements on the IP transition “break the Internet?”

Remember the remarkable near unanimity in the U.S. Government, Congress, industry, and NGOs opposing the International Telecommunications Union (ITU) intention to assert regulatory control over the Internet?

Well if the FCC imposed interconnection regulatory requirements on the Internet in the U.S., every country in the world seeking to control or censor the Internet in their country, could claim that they were just following the example of the U.S. in regulating how the Internet works for economic and competitive reasons.

Simply, every despot seeking political cover for censoring its populace’s use of the Internet could claim that their imposition of interconnection regulation on all Internet traffic in or out of their country or province was for economic and competitive reasons not political ones.

In sum, setting the precedent for interconnection regulation of the Internet would “break the Internet” in three devastating ways.

First, it would lead to a balkanization of the global Internet into many artificially-created national or provincial sub-Internets by facilitating the imposition of sovereign border interconnection checkpoints where each nation or province could devise their own Internet border traffic rules for the first time.

Second, it would break the current bottom-up, multi-stakeholder, technological governance structure of the Internet by imposing an alien, top-down, political-economic, governance structure on geographic subdivisions of the Internet.

Third, there is no better way to break the Internet than politically forcing the Internet’s inherently discontinuous, decentralized, and modern, Internet packet delivery system to operate antithetically to the Internet’s design and architecture – i.e. like a continuous, centralized, obsolete telephone system.

That’s why IP interconnection would break the Internet. Forewarned is forearmed.

Scott Cleland is Chairman of NetCompetition a pro-competition e-forum supported by broadband interests and President of Precursor LLC, a research consultancy for Fortune 500 companies. Cleland served as Deputy U.S. Coordinator for International Communications and Information policy in the George H.W. Bush Administration.