> U.S. intelligence officers were also able to identify digital links between the covert communications system and the U.S. government itself, according to one former official—links the Chinese agencies almost certainly found as well. These digital links would have made it relatively easy for China to deduce that the covert communications system was being used by the CIA. In fact, some of these links pointed back to parts of the CIA’s own website, according to the former official.
How was this approved? This is such an obvious no-no. Would have expected a completely standalone, ring-fenced platform with absolutely no relationship to the government's known IP addresses and domains?
How do we know the CIA didn't kill Kennedy?
He's dead isn't he?
If it's off the shelf and others are using, then they can just buy the equipment and look for flaws (like this massive one).
Threat model from China in terms of infosec is very different than middle East countries (except isreal perhaps).
Not that it was a good idea in either region.
And who do you think developed this communications system they use? Given government pay scales probably relatively junior web developers trained and used to a consumer-oriented software culture in which "oh let's just drop in this convenient re-usable library" is a no-brainer.
I really don't think there's anything to conclude from the "firewall" comment. This term has become a layman way of describing any kind of intended separation between computer systems. People know what firewalls are and so it becomes an overused analogy. It's like how passwords are described as being "encrypted" in layman media.
I would bet that communication apps A and B made connections to the same someCIADomain/IP range (or other less obvious metadata patterns, maybe the shared methods produced enough entropy on the whole to fingerprint).
Thus even if one can't read the content of the message the firewall matches the discovered pattern === likely a spy; so if the less vetted App A source was compromised offline and thus the shared online pattern is now known so too are sources using App B who were less likely to be compromised offline.
Compare that to one way coded messages over hf radio: a place for pick up of a dead drop in n days is broadcasted, then it is picked up by a man who reads it, destroys it, and gtfos from the country. In that scheme nobody in any way contacts anybody in real time other that the HQ.
Having a shortwave radio in your possession is justification enough to haul you in for questing in some countries.
There is also relatively little "radio traffic" to inspect relative to internet traffic.
If it ain't broke, don't fix it?!
No, they've always had both roles. For instance, their experience and knowledge of advanced code-breaking techniques allowed them to strengthen the DES cipher's S-boxes in ways that weren't publicly understood until decades later.
https://en.wikipedia.org/wiki/Data_Encryption_Standard#NSA%2...
This is naught but my opinion, and I understand where others might differ in their reading of the situation presented to us.
Deploys on computers. Web-based. Detectable through decryption or pattern analysis. Firewalls. It was a pair of Tor hidden services. The mistake was probably that they ran both on the same machine with only a firewall separating them, rather than physically different servers.
On the other hand I'm surprised tor works in China at all. I would expect the Great Firewall to shut down all entrance/exit nodes going into or out of the country, except maybe for a few controlled by the government.
Oh look at this cute kitten picture, let me save it and then get the encrypted code out of it. That's great, let me upload this equally cute puppy riding a skateboard picture to the forum that contains my reply.
>In the words of one of the former officials, the CIA had “fucked up the firewall” between the two systems.
If you read between the lines, this raises the suspicion that there's a common underlying infrastructure which handles the communications, with management front-ends for different users which are firewalled off from each other, and the security of the system relied upon the firewall between the different front-ends to prevent users from finding out about each other. However, an attacker who compromised the "less secure" front end, could use that as a launching pad to attack the underlying communication infrastructure, and if the attacker pwned the infrastructure, then he'd have a back entrance to the "more secure" front end.
If that's the case, then somebody was grossly incompetent, depending on the age of the system: if the system is old enough, then somebody running ops in the CIA is incompetent, for continuing to operate a system whose security model ("all you need is a strong enough firewall!") was obsolete; if the system is young enough, then either the original architects, or the security engineers who certified the architecture (if there were any), for proposing an architecture with an obsolete security model.
Arguably, that incompetence amounts to criminal negligence, since it resulted in the deaths of US agents, and somebody should be tried for it.