If you could re-invent the entire Internet, starting all over again with security in mind, what would you do?



Check out this post for the discussion that is the basis of our conversation on this week’s episode co-hosted by me, David Spark (@dspark), the creator of CISO Series and Allan Alford (@AllanAlfordinTX). Our guest for this episode, Davi Ottenhimer (@daviottenheimer), who is working at Inrupt, Inc with Tim Berners-Lee on reinventing the web.

Got feedback? Join the conversation on LinkedIn.

Thanks to this week’s podcast sponsor, Castle

Castle is helping businesses keep customers’ online accounts safe from targeted account takeovers, automated credential stuffing, and risky user transactions. Castle’s user-centric approach to account security allows organizations to fully automate threat response and account recovery in real-time with risk-based authentication, granular access policies, and custom workflows. Learn more at www.castle.io

On this episode of Defense in Depth, you’ll learn:

  • Much of the advice on how to secure the Internet focused on just improving known protocols such as SMTP, IPv6, and TCP/IP. Is that limited thinking or not?
  • Creating a new Internet has a lot of political and socioeconomic issues connected to it so you have to consider both relative (changing existing protocols) or absolute updates (reinventing and trashing existing protocols).
  • One suggestion was dynamic port assignments which was an interesting tip, but it runs into the issue that at some point someone needs to know where you’re communicating.
  • Future of identity is that it’s not controlled by one entity. But the solution is not blockchain. That’s essentially a spreadsheet of information and banking on a spreadsheet or blockchain would not be wise.
  • Another suggestion would be to create a data-centric approach to the Internet, but this would put a massive load on the endpoints.
  • One core philosophy of securing the new Internet is creating a system where each individual can own their own data, put rights on it to others to use it, rather than being beholden to the rights others give us to manage our own data.
  • Our favorite suggestion was about looking to biomimicry and our millions of years of evolution to help us build an Internet that could learn to evolve on its own. The issue is that history has given us tectonic shifts that come all at once and don’t necessarily evolve gradually. Could a security system be built to adapt in that manner?

Creative Commons photo attribution to Joybot.