News for hackers. Stuff that matters.
Re:OnlyKey: Open-Source Alternative to YubiKey by userbinator on Thursday December 26, 2019 @10:07PM (#21888819)
I can safely say that a lot of proprietary crypto code (as in, stuff that is in very widespread use and costs $$$) is not unlike this either. In some ways this is actually more straightforward to read and understand since it's in one file and not wrapped in a dozen layers of abstraction.
Re:OnlyKey: Open-Source Alternative to YubiKey by cr7pt0 on Friday December 27, 2019 @09:39AM (#21891541)
Definitely true, anyone who has ever seen proprietary crypto code knows this. Reviewing one file that is 7000 lines long is more straight forward than reviewing 7000 lines of code split in multiple files. It's open source and we will continue to make it better. If the biggest criticism here is the large file size, RNG complaint (top post is incorrect about analog read, they missed that we also use 6 touch buttons to seed RNG), and code style then it's a safe bet that OnlyKey source is better than most of the proprietary security keys out there. Of course it's not possible to know for sure as they are closed source, but you can look at past vulnerabilities. Like this one https://crocs.fi.muni.cz/public/papers/rsa_ccs17 it's not a theoretical RNG issue like the criticism here has been, it's an actual exploitable vulnerability that affected Yubikey and tons of smart cards. This exploit was on devices that were already FIPS and CC certified. Another thing to consider is the way the researchers found this was by statistically testing a bunch of keys, they didn't even review the source so you can imagine how many more security vulnerabilities they would find if they did.