Rostislav Svoboda <[email protected]> writes: > What I call for is the abolition of a protection mechanism that fails > to protect us, only leading us into a false sense of security. > > Again, it's not about who wrote the code - it's about what is in the code.
Sure. But nobody claims that 'guix authenticate' protect us against bad code, or? The claim, which I believe is correct, is that 'guix authenticate' provides some security assurances that git's normal SHA1-based merkle design and PGP signed git commits does not. I worry that even 'guix authenticate' is not strong enough, and that its limits aren't well understood. I would worry more if we went back to a trust-SHA1-and-forge-admins-and-the-webpki-private-key-owners mode, which fail to address reasonable security concerns on all three trust aspects. /Simon
signature.asc
Description: PGP signature
