[powerpress]

Cybersecurity is on the front burner of Internet concerns for both users and policy makers thanks to relentless, high profile attacks. We’ve all heard about data breaches at Sony, OPM, Ashley Madison, Experian/T-Mobile, CVS, and the IRS, but Hackmageddon lists 30 data breaches in the month of July alone. While some of our brightest security stars are working on problem, they’re essentially bailing water. Is something wrong with the boat?

The Internet we have today is the offspring of a research project whose Statement of Work called for a means of exchanging large data sets within a trusted community of researchers and academics, but that’s not what we need today. One of the more radical ideas for dealing with security and a host of other issues affecting the Internet is to redesign the Internet’s architecture. The argument for a new Internet architecture is made very well in John Day’s Patterns in Network Architecture: A Return to Fundamentals, which argues that the Internet went down the wrong path when TCP was divided from IP and the INWG 96 consensus design was rejected by ARPA.

On this week’s High Tech Forum podcast, Professor Day joins Richard Bennett for a discussion Day’s alternative Internet, RINA (the Recursive Internet Architecture.) Day lays out the elements of RINA at a high level, discusses the nature of the RINA transition, and assesses the state of three large-scale RINA trials in Europe.

RINA achieves security by design rather than through the post add-ons to the legacy Internet that have made the security breaches with which we’re all-too-familiar. It’s a fresh new approach to an old problem that shows enormous promise, a complement to Software-Defined Networks and other means of compartmentalizing high-value information away from prying eyes. Professor Day provides additional reading in this essay and this presentation.

Highlights of the podcast:

• The early Internet had poor security because computers in the 1970s were weak: “you can buy toys at Toys ‘R Us right now that have more computing power than we had in the ARPANET era.”
• Security has become so expensive that some analyses suggest that we spend more on attack mitigation each year than we gain in productivity from using the Internet: “…the cost-benefit ratio of the Internet has become negative because of the costs of security.”
• Network theory hasn’t kept up with practice because universities aren’t teaching the necessary skills: “…we’ve been selecting, for the last twenty-five years for craftsmen in the field of computer science rather than theoreticians.”
• According to Bob Metcalfe, the inventor of the early Ethernet networking is inter-process communication, similar to functions that take place within computer operating systems.
• The transition from old Internet architecture to RINA is a matter of adoption: “…it’s not transition, it’s adoption. We can run RINA over IP, we can run RINA under IP.”
• RINA’s design is similar to that of Wi-Fi, with an enrollment phase before data transfer can begin.
• Security by design means application developers can ignore security: “…actually what we have with RENA is security by ignoring it.”

Join us for a fascinating discussion of the Internet’s future with a veteran of the Internet’s earliest origins