News

Today’s episode is about the new version of the TLS protocol, because recently first Chrome https://www.chromestatus.com/feature/5712755738804224 and now Firefox https://www.bleepingcomputer.com/news/security/mozilla-has-started-gradually-enabling-tls-13-in-firefox/, https://bugzilla.mozilla.org/show_bug.cgi?id=1310516 have enabled it by default, and at the same time its RFC has been finally approved for Standards track mode https://www.ietf.org/mail-archive/web/ietf-announce/current/msg17592.html. This is a big thing that has been waiting for an awfully long time (on the server side, production support started to appear for example in large CDNs more than a year ago https://blog.cloudflare.com/why-tls-1-3-isnt-in-browsers-yet/) and it is interesting to look at the reasons for these peripeteia that lie in the considerable controversy of the new protocol version.

At the heart of the controversy is that TLS 1.3’s features (e.g., https://blog.cloudflare.com/introducing-0-rtt/, which reverses the delay ratio of encrypted to unencrypted connections) strongly motivate an increase in both the share and security of encrypted communication. Together with the new version of HTTP 2, which requires an encrypted transport layer protocol, and the non-profit activity of the free certificate provider Let’s encrypt, it is likely to lead the Internet into a new era in which encryption of communications should become commonplace for even the most trivial sites and servers. This might seem like an exclusively positive thing, but it very much hits the little-known grey area of corporate proxies and other methods of deep inspection of network traffic.

Indeed, there is a lot of software that has exploited the shortcomings of previous versions of TLS to closely monitor and analyze even encrypted network traffic in controlled environments (typically corporations and government institutions, but also telephone carrier networks, for example). The problem is that these procedures were de facto identical to those used by hackers to launch attacks, which has led to a situation where it is not always easy to tell exactly who is still a legitimate administrator and who is already an illegal aggressor. This state of affairs has somehow been quietly overlooked until now, but TLS 1.3 addresses these shortcomings, forcing the professional community to explicitly answer the highly political question of how much privacy and personal integrity we are willing to sacrifice for collective security.

The adoption of TLS 1.3 marks a significant tilt towards the privacy side in this age-old dispute, and it is this controversy that has caused the many years of delay in the preparation of TLS 1.3

Author: František Řezáč