Confidentiality in the Face of Pervasive Surveillance
datatracker.ietf.org
datatracker.ietf.org
The kind of capital-F Freedom that is supposed to exist in the Free World is in practice just the gap between what's practical to control hierarchically and the level of desired power to control (effectively limitless). Technology continues to narrow the gap.
Some of the themes of societal change in my lifetime are: reduced trust, reduced privacy, centralization of power, increased economic efficiency at the expense of robustness, and increased profit extraction at the expense of nearly everything else.
I have all of this scripted to launch trivially within a few seconds, and it reduces the effectiveness of most of the attacks this RFC describes.
Echelon was known by the late 1990s and has now been running for at least a third of a century. The European Parliament report came out in '01. Snowden and co, which this IETF document was presumably issued in response to, was a decade and a half later, a time in which offensive population-scale surveillance and targeted penetration as a service became open market products, secret global state-sanctioned kidnappings and assassinations became nominally normalized, literal constellations of surveillance satellites have been launched at ever-decreasing cost with ever-increasing capabilities, US 'instant global strike' capability is declared, drone warfare has not only been realized but is constant and in the popular consciousness, and every state actor worth its stazi-salt started playing.
IMHO this document is therefore primarily of historical interest in demonstrating the apparent, unfortunate, well-meaning near-irrelevance of the IETF in the face of global internet commercialization and the ongoing shift toward normalized state surveillance. It could be seen as the conscience of the internet coming to terms with its failure. In Assange's words of 2012 (some 3 years prior): "The internet, our greatest tool of emancipation, has been transformed into the most dangerous facilitator of totalitarianism we have ever seen. The internet is a threat to human civilization." This is all the more significant in mid 2024 as Assange has just escaped torture, China and Russia tighten their grasp after abandoning any open market pretense, Europe, at war, stands on the cusp of a popular authoritarian transition and the US election looms.
What's a regular citizen to do? A shell script certainly won't save you, but it's something. It seems humanity has some organizational problems - its political systems no longer seem fit for purpose, and all of this while biodiversity tanks and global warming accelerates. Post industrial revolution we've built a machine, and it's no longer under our control. To Mars then? Perhaps not.
Regardless I'm not really trying to defend myself against state actors. If they target me, I know I'm screwed. What I am trying to do is reduce the amount of my private information non-government entities collect on me while raising the cost of mass surveillance.
Maybe physical torture.
Was hoping for some new tips on what to do about government agencies demanding information from constituents and then storing it in MS Windows.
But yes, Windows does it on a whole new level.
I for one propose an unwritten language of moans and yodels steganographically modulating a subset of classical Chinese atop boudoir French poetry, preferably set to sitar.
2) Therefore: "Protocols that do not encrypt their payload make the entire content of the communication available to the idealized attacker along their path."
3) Furthermore: "When store-and-forward protocols are used, intermediaries leave this data subject to observation by an attacker that has compromised these intermediaries..."
I will now prove a negative: If the idealized attacker has 1 then they've already achieved 3. Therefore this is a bullshit line of argument.
What next? Read your eyeballs with your web cam? Do I need to encrypt traffic on loopback? Where is "reasonable" here? There is no such thing as absolute zero trust or absolute air gap. Is the "idealized defender" an idiot who thinks that encrypting loopback mitigates pwnage at the os or hardware level? What role does deception, salting, misdirection play in a healthy security posture?
This should be interpreted as a followon to RFC 7258 (cited in the Introduction): "While PM is an attack, other forms of monitoring that might fit the definition of PM can be beneficial and not part of any attack, e.g., network management functions monitor packets or flows..."
I am reminded of this from RFC 1034: "Clients of the domain system should be able to identify trusted name servers they prefer to use before accepting referrals to name servers outside of this 'trusted' set." and rhetorically now I ask how has that worked out?
Not discounting that the technical issues and scenarios are valid, but to point out the blatantly political nature of this rhetoric.
What this gets us at the level of technical implementation is e.g. qname minimization, which in turn begets moaning about lame delegations (I don't deny they are a problem) and the horrible horrible excess traffic they cause... entirely omitting the fact that qname minimization can double the number of unprimed queries required to resolve a name. Just one example.
Not in somebody else's cloud. :-p
But ok. So there is the issue of compilers carrying binary-only payloads, which are only passed on when compiling the compiler. Really paranoid people worry about this, so it fits in the set of theoretical and demonstrated issues. But I don't have references for you off the top of my head; it's been acknowledged for a very long time.
I'm ok with heterogenenaety and calling it good enough.
Open source could write 5 fully functional and modern Software Development, Distribution and Running Infrastructures (currently known as an OS) per day if all the effort didn't went into the hospitalization of the current ones and the ecosystems surrounding them.
In the spirit of debate only, I observe that the Theseus legend concerns the farmer's axe: replaced the handle three times and the head twice, still a damned good axe.
Thus marches whatever software you cite: the legionnaires may die and be replaced but the legion marches on. A lot of the burgeoning cost concerns camp followers rather than the legion itself. You seem to get that in your reference to ecosystems. Having trouble getting water in camp? If your pack animals are dying it's a different concern than your follower's. I mean, whose animals are you going to eat first?
I would like nothing better than to be running on a kernel written in 1584. (Arbitrarily chosen year, seriously.)
Qubes OS comes to mind.