Reply to post: Re: Yes but

'Trust it': Results of Signal's first formal crypto analysis are in

Adam 1

Re: Yes but

The two statements that concern me about this research are:

1. Signal employs a novel and unstudied design, involving over ten different types of keys and a complex update process which leads to various chains of related keys

Novelty is not a positive feature. It doesn't necessarily mean it's negative (all designs were at some point in human history considered novel in this sense) but anything that makes it harder to study is just security through obscurity. In the same way obscurity doesn't mean insecure, but the obscurity may mask some actual flaws from the whitehats/design reviewers so the security ends up compromised.

That leads to

2. the protocol is not substantially documented beyond its source code

Given the supposed advantage of the novel design, the design itself should be will documented at a high level so that inherent design flaws can be effectively studied. Not the implementation itself (through implementation bugs also need to be checked) but the interaction between the parties with data/keys/RNG etc for inherent attack vectors.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

SUBSCRIBE TO OUR WEEKLY TECH NEWSLETTER

Biting the hand that feeds IT © 1998–2021