If you have evidence, I'm sure you can bring to the attention of the Mozilla Root Store Inclusion program and the CA/Browser forum. Moreover, "there are other criminals who have gotten away with much more" is not an argument.
I think the steel man version of the argument is: "You showed that there is no effective monitoring or transparency that me as a user can get, and as such, Where is the trust come from. What fundamentally causes Mozilla (or any) to trust CAs more than randomly distributed certificates."
The CA/Browser forum's requirement and enforcements such as this one (and against DarkMatter, CNNIC and the likes) give me the required confidence to trust them, even though I'd agree it's not a perfect process.
Your average user is unlikely to begin to understand why a CA would be trustworthy, and a web of trust model only works for social situations but not for certificate distribution.
"A moose once bit my sister. Therefore all meese must be sacked".
I have trouble seeing this as a steeled version of anything. "People have uncovered a flaw in the system, therefore the entire system is unfit for purpose" does not really make a compelling argument. It displays selection bias, hasty generalization, nirvana fallacy, and something about babies and bathwater.
No I don't have evidence - however logical deduction shows that the probability of this happening is high. Any system involving humans is fallible, so it would be naïve to think that it doesn't happen.
Or put another way: if I was the NSA or MI5 this is exactly how I would attack the problem of traffic interception or targeted black ops. Get a puppet CA via hook or crook.
Totally agree that "there are other criminals who have gotten away with much more" is not an argument; I'm not sure what that has to do with my comment? I'm certainly not suggesting that. If anything I suggest that such systems are pretty much broken by design (at least if you care about state actors / extremely well funded actors).
yonixw|3 years ago
stevewatson301|3 years ago
Your average user is unlikely to begin to understand why a CA would be trustworthy, and a web of trust model only works for social situations but not for certificate distribution.
tremon|3 years ago
I have trouble seeing this as a steeled version of anything. "People have uncovered a flaw in the system, therefore the entire system is unfit for purpose" does not really make a compelling argument. It displays selection bias, hasty generalization, nirvana fallacy, and something about babies and bathwater.
pas|3 years ago
I know nobody likes DNSSEC, but DANE works too :)
radicalbyte|3 years ago
Or put another way: if I was the NSA or MI5 this is exactly how I would attack the problem of traffic interception or targeted black ops. Get a puppet CA via hook or crook.
Totally agree that "there are other criminals who have gotten away with much more" is not an argument; I'm not sure what that has to do with my comment? I'm certainly not suggesting that. If anything I suggest that such systems are pretty much broken by design (at least if you care about state actors / extremely well funded actors).