alt Hacker NewsAgain: which experts were saying what was impossible, which was then revealed to be possible by the Snowden documents?
Is the claim that there was adequate court oversight of operations under those codenames which then turned out not to be the case? Are they referring to specific excesses of the agencies? Breaking certain cryptographic primitives presumed to be secure?
Why is absolutely no one who knows all about Snowden ever able to refer to the files with anything more then a bunch of titles, and when they deign to provide a link also refuses to explain what part of it they are reacting to or what they think it means - you know, normal human communication stuff?
(I mean I know why, it's because at the time HN wound itself up on "the NSA has definitely cracked TLS" and the source was an out of context slide about the ability to monitor decrypted traffic after TLS termination - maybe, because actually it was one extremely information sparse internal briefing slide. But boy were people super confident they knew exactly what it meant, in a way which extends to discussion and reference to every other part of the files in my experience).
I mostly focused on the cryptographic parts of the files. Here's what I wrote after the first details of cryptographic attacks were released: https://blog.cryptographyengineering.com/2013/09/06/on-nsa/
What I learned in that revelation was that the NSA was deliberately tampering with the design of products and standards to make them more vulnerable to NOBUS decryption. This surprised everyone I knew at the time, because we (perhaps naively) thought this was out of bounds. Google "SIGINT Enabling" and "Bullrun".
But there were many other revelations demonstrating large scale surveillance. One we saw involved monitoring the Google infra by tapping inter-DC fiber connections after SSL was added. Google MUSCULAR, or "SSL added and removed here". We also saw projects to tap unencrypted messaging services and read every message sent. This was "surprising" because it was indiscriminate and large-scale. No doubt these projects (over a decade old) have accelerated in the meantime.