|
Very cool! When designing psst, I considered implementing it as a self-contained website. The advantages would be (1) ease of use, and (2) more features. In contrast, psst is a simpler design, so simple that the entire system fits on two pages of paper. I never quite managed to make a self-contained website that was small enough to understand/print/etc. Thus, psst is targeted towards people who want a system that they can fully understand and verify, whereas s4 is easier to use. I was also worried that people might end up with enough shares, but no possibility to reconstruct them. With psst, shares are fully self-contained. A user with two shares has all they need to obtain the secret, plus typically some context that tells them what to do with the secret. With s4, this is a bit more tricky: imagine your heirs finding some papers with s4 shares; they would have to find the s4 website first. Maybe your domain name has since expired and they need to dig a copy out of archive.org... Have you thought about ways to solve this? For example, it would be nice if s4 could generate zip files containing a share, and all the files needed to run the s4 decoder. (Assuming that zip, JavaScript, and WebAssembly are still easily usable in 2044.) |