|
Today marks the 20th anniversary of the decommissioning of the NSFNET backbone on April 30 1995, an important milestone in the development of the commercial Internet. The NSFNET was set up by the US National Science Foundation in 1985 to enable university researchers access to five supercomputer sites across the United States, using Internet Protocol technology. In stepping back, the NSF supported a transition to an Internet shaped by market forces, and the explosion of commercial use soon followed.
The changes since that time are truly breathtaking and bear reflection. In 1995 there were 16 million Internet users—now there are likely over 3 billion; in 1995 there were under 5 million hosts; now there are over 1 billion; and the NSFNET had 45 Mbit/s capacity across the network—a bandwidth level that can be exceeded by a single mobile broadband access connection today.
It is impossible to determine what would have happened if the NSFNET had not been decommissioned. Netscape had just released its first browser, which no doubt would have led to an explosion of World Wide Web traffic for commercial backbones regardless. However, there seems little doubt that the way that the NSFNET functions were replaced, contributed greatly to the vibrancy and growth of the Internet through today.
As it happens, my first exposure to Internet issues came in 1997 when I was at the US Federal Communications Commission and led the review of Internet backbone issues arising from the MCI WorldCom merger, and then later from the Bell Atlantic GTE merger. I was struck by the lack of any interconnection regulations on Internet services at the same time that telecommunications services, using the same or similar facilities, were regulated.
This led me to write a working paper entitled ‘The Digital Handshake: Connecting Internet Backbones’ (FCC Office of Plans and Policy Working Paper 32, 2000) to examine how interconnection was reached in the absence of any regulatory obligations. I soon learned that the foundation for the lack of any interconnection regulation was laid by the NSF, which fed into the FCC’s own position of not regulating Internet services at the time.
The NSFNET was a backbone in the true sense of the word—the spine through which an increasing number of regional educational and research networks connected to each other, and the supercomputers. The few commercial networks at the time, exchanged traffic through the NSFNET to the extent allowed by the NSF acceptable use policy limiting commercial traffic, and setup the first commercial exchange in 1991 to directly exchange the rest of their traffic.
As such, a significant issue that had to be addressed was what would take the role of the NSFNET in exchanging traffic, and the answer was that the NSF awarded contracts to manage and operate four privately owned network access points (NAPs) scattered around the US, where providers could exchange traffic with one another. While establishing the NAPs, the NSF notably did not impose any rules or regulations as to how and with whom traffic was exchanged.
In this vacuum, commercially negotiated peering and transit arrangements took hold. These arrangements remain a steady constant today against a backdrop of unending change since that time:
The NSFNET, and the NAPs that originally replaced it, are now an anachronism in an Internet in which the US no longer occupies its historical central role as founder, but whose founding principles expand with the Internet to every corner of the World. This is nowhere more true than here in Kyrgyzstan, where I am this week with a team from the Internet Society, working with the local ecosystem to improve Internet interconnection arrangements, and thereby help to initiate for the country the same evolution and growth that the NSF helped kick off for the world 20 years ago.
This post originally appeared on the Internet Society’s blog
Sponsored byVerisign
Sponsored byIPv4.Global
Sponsored byWhoisXML API
Sponsored byCSC
Sponsored byRadix
Sponsored byVerisign
Sponsored byDNIB.com