4 ms·

Doesn't really undermine your point, but as a fun aside: Trusting Trust style attacks on compilers were (theoretically) solved by David Wheeler with "diverse do

• caust1c 2y ago
  Yes, good point. That's pretty similar to scientific falsifiability at least! I think for topics that are not as easily provable as reproducible builds though, trust gets murky. There was something else I read (can't find it now) that made a similar analogy for a web of trust. A web of trust (e.g. PGP) will have de-facto authorities since there will be a tendency for more people to sign individuals presumed to be trustworthy based on their history. It follows that the system runs into issues if their key gets compromised or if a false individual is subject to a sybil attack, producing the illusion of trust. See also: github stars, cryptocurrency, social media follower counts.