4 ms·

"My scheme was covert communications embedded in HTTPS to blend in, remailers through various jurisdictions that don't cooperate, face-to-face for most stuff, d

"My scheme was covert communications embedded in HTTPS to blend in, remailers through various jurisdictions that don't cooperate, face-to-face for most stuff, dead drops, calls with lineman's headset, snail mail of CD's surrounded by coupons etc, and so on. Stuff that draws no attention or has strong security against likely threat profile. Not sure how much of it would work today but no tech and low tech methods are still best." This is way off the mark I think, how can something that requires _physical presence_ be best? Bringing up Bin Laden is not a good example here, as the particulars of his case made using cyber extremely difficult. The Internet is absolutely and by far the best way for anonymous and/or covert communications between individuals or groups, as long as they possess domain expertise. And this is only going to get better, not worse, mass-surveillance notwithstanding, due to the proliferation of information. That doesn't necessarily mean everyone has to use TOR and PGP, just that it's very easy to _hide_ amidst the noise and not compromise on OPSEC. Lots of ways to do that.

nickpsecurity 10y ago
How can it be way off the mark if such methods have worked for decades while online communities keep getting decloaked and secrets inisibly copied? I think what already worked should be baseline go compare against. Plus, you can use Tor within my model so long as it's an untrusted layer among others. Important they dont see you using it. Bin Laden was a good example because he likewise employed good, pre-Internet OPSEC. He stayed away from electronic communications mediums, used physical medium for data storage, moved it with trusted couriers, and couriers activity looked normal to observers. These principles are compatible with my recommendations and held off the most powerful, SIGINT attacker in existence. They can work for others, too. ;) "That doesn't necessarily mean everyone has to use TOR and PGP, just that it's very easy to _hide_ amidst the noise and not compromise on OPSEC." Snowden links directly contradict you. Such people and data stick out like a sore thunb. They also collect what encrypted data tgeg can for five years. People using my methods would likely have been filtered and ignored. Or, avoiding the net, they'd have never been seen at all.
- armitron 10y ago
  You are making wide proclamations and generalizations with nothing to back them up, this is not the mark of someone with a deep understanding of the subject matter. You're also all over the place, rather than concentrating on the point I made [physical presence]. What worked for decades, for some definition of worked -> debacles are numerous [when the Internet didn't even exist or wasn't as widely deployed/used as it is now] doesn't mean it is best _today_. This is a fool's argument, also see what Taleb has to say about tail events. Moreover, the OPSEC details of most successful groups and individuals that use the Internet to communicate, will by definition, rarely become public. We can learn by example from failures but we can't learn by example from success _unless we are insiders_. I said the particulars in the Bin Laden case are important and made cyber hard to use. That doesn't mean trusted couriers and physical storage are a good idea for the rest of us. Re: Snowden You yourself said you were using comms embedded in HTTPS to avoid sticking out like a sore thumb, how do Snowden docs contradict what I said? Read [and try to comprehend] my post again please. To reiterate, I think domain expertise gives individuals a certain degree of confidence and a solid knowledge base that they can use to build their OPSEC strategies. What would you gamble your life on if you had to, meatspace domain expertise [with a huge number of totally unknown factors/capabilities] or Internet [some unknown factors but also lots of remedies]? For most of us the answer is obvious. When I go out, I can never be confident I'm not being followed, recorded in various ways or leaking biometric evidence. Online, I can operate with an assumption of full compromise _and still come up with useful OPSEC strategies_.
  - nickpsecurity 10y ago
    " rather than concentrating on the point I made [physical presence]" A method for surveilling people's activities ideally has these properties: 1. Delivers intel on as many targets as possible. 2. Requires little to no effort on operator's part. 3. Is easy to deploy multiple times or stays persistently. 4. Doesn't reveal presence of surveillance. 5. Takes a lot of work to avoid. 6. Has reasonable cost. A method for concealing communications or activities has at least these properties in this order (except for 5 which always helps): 1. Prevents detection of activity entirely against likely eavesdropping methods in use. 2. Makes the eavesdropping difficult to attempt 2-1. Focused instead of mass surveillance. 2-2. Expensive. 2-3. Physical. 2-4. Requires talent instead of brainless drones. 3. Increases odds of person being surveilled detecting the surveillance. 4. Allow for countermeasures to end or recover from the surveillance. 5. Is deniable and inexpensive. So, let's look at average person using common recommendations. The average person will be using Torbrowser, TAILS, Windows + Tor, Mac + Tor, Linux desktop + Tor, or main mobile OS + Tor. They will not be professionals at attacking or securing systems like the criminal or government hackers tasked with uncloaking them. The attacks on Tor, by NSA & by academics, apply where many users get unmasked remotely in a way they never see. They can all be surveilled at once by NSA's existing infrastructure. Outside this, the FBI deanonymized users by using a browser 0-day (endpoint attack). Five Eyes, esp NSA, have 0-days on each desktop and mobile OS with high, market share per leaks. We regularly see Russian and Chinese hackers breach them in the news. We also see all kinds of hackers breach systems in general plus lots of bad decisions by users. The risk of a stealthy, mass, inexpensive leak of people using digital communications is very high whereas the odds of them outthinking average surveillance is low and targeted hacking very low. So, let's look at people using meatspace with methods like I described above and simple guides on their use. They've spent their whole life learning how to tell when people are watching them, nervous, or just acting unusual. They have numerous ways of blending into a crowd where WiFi or drops are available but with limited camera coverage. They can learn the traffic patterns of the area to see if anyone sticks out. Likewise, there's private spots to stash a long-range antenna and camera to do the same from a distance. Many know how to ditch evidence or pay for stuff with cash. Keeping their name and I.P. out of it reduces odds they'll be seen where no tech or HTTPS-style comms increases odds surveillance ignores them entirely as mere Internet, background noise. These methods were and are used successfully by many crooks and spies... today just like decades ago with drops, radios, and countersurveillance... while "smart" people, their OS's, and anonymity software get owned on computers quite regularly. So, the average person looks to be better off avoiding the Internet, executable media, or anything that draws attention when trying concealed communication with other average people. It's easier for them to follow this advice like that than learn arcane, technical information that changes constantly. It also turns inexpensive, stealthy, mass attacks into requiring footsoliders focused on one person or a small group at a time risking detection constantly. That's also with no guarantee they'll keep whatever gains they acquire in intelligence collection. That's a significant increase in effectiveness versus new methods where you trust things that can be and are often invisibly hacked by remote attackers. "When I go out, I can never be confident I'm not being followed, recorded in various ways or leaking biometric evidence. Online, I can operate with an assumption of full compromise _and still come up with useful OPSEC strategies_." No you can't. If you can't trust or mitigate physical risks, then you can be compromised in the physical realm to defeat your software countermeasures. There's whole Snowden slides dedicated to those things. Not to mention books by experts in hacking, black bag jobs (esp TSCM), infiltration (esp spec ops), and crime (esp B&E's or cons). Physical security and lack of subversion is a prerequisite for and/or major part of online security. So, what you're really doing is hoping you're not worth their time or money at all plus no flaws are found in your hardware, firmware, OS, browser, Tor, or any other running software. Might be true but lots of counterevidence on software side. ;)

nickpsecurity 10y ago

How can it be way off the mark if such methods have worked for decades while online communities keep getting decloaked and secrets inisibly copied? I think what already worked should be baseline go compare against. Plus, you can use Tor within my model so long as it's an untrusted layer among others. Important they dont see you using it. Bin Laden was a good example because he likewise employed good, pre-Internet OPSEC. He stayed away from electronic communications mediums, used physical medium for data storage, moved it with trusted couriers, and couriers activity looked normal to observers. These principles are compatible with my recommendations and held off the most powerful, SIGINT attacker in existence. They can work for others, too. ;) "That doesn't necessarily mean everyone has to use TOR and PGP, just that it's very easy to _hide_ amidst the noise and not compromise on OPSEC." Snowden links directly contradict you. Such people and data stick out like a sore thunb. They also collect what encrypted data tgeg can for five years. People using my methods would likely have been filtered and ignored. Or, avoiding the net, they'd have never been seen at all.

armitron 10y ago
You are making wide proclamations and generalizations with nothing to back them up, this is not the mark of someone with a deep understanding of the subject matter. You're also all over the place, rather than concentrating on the point I made [physical presence]. What worked for decades, for some definition of worked -> debacles are numerous [when the Internet didn't even exist or wasn't as widely deployed/used as it is now] doesn't mean it is best _today_. This is a fool's argument, also see what Taleb has to say about tail events. Moreover, the OPSEC details of most successful groups and individuals that use the Internet to communicate, will by definition, rarely become public. We can learn by example from failures but we can't learn by example from success _unless we are insiders_. I said the particulars in the Bin Laden case are important and made cyber hard to use. That doesn't mean trusted couriers and physical storage are a good idea for the rest of us. Re: Snowden You yourself said you were using comms embedded in HTTPS to avoid sticking out like a sore thumb, how do Snowden docs contradict what I said? Read [and try to comprehend] my post again please. To reiterate, I think domain expertise gives individuals a certain degree of confidence and a solid knowledge base that they can use to build their OPSEC strategies. What would you gamble your life on if you had to, meatspace domain expertise [with a huge number of totally unknown factors/capabilities] or Internet [some unknown factors but also lots of remedies]? For most of us the answer is obvious. When I go out, I can never be confident I'm not being followed, recorded in various ways or leaking biometric evidence. Online, I can operate with an assumption of full compromise _and still come up with useful OPSEC strategies_.
- nickpsecurity 10y ago
  " rather than concentrating on the point I made [physical presence]" A method for surveilling people's activities ideally has these properties: 1. Delivers intel on as many targets as possible. 2. Requires little to no effort on operator's part. 3. Is easy to deploy multiple times or stays persistently. 4. Doesn't reveal presence of surveillance. 5. Takes a lot of work to avoid. 6. Has reasonable cost. A method for concealing communications or activities has at least these properties in this order (except for 5 which always helps): 1. Prevents detection of activity entirely against likely eavesdropping methods in use. 2. Makes the eavesdropping difficult to attempt 2-1. Focused instead of mass surveillance. 2-2. Expensive. 2-3. Physical. 2-4. Requires talent instead of brainless drones. 3. Increases odds of person being surveilled detecting the surveillance. 4. Allow for countermeasures to end or recover from the surveillance. 5. Is deniable and inexpensive. So, let's look at average person using common recommendations. The average person will be using Torbrowser, TAILS, Windows + Tor, Mac + Tor, Linux desktop + Tor, or main mobile OS + Tor. They will not be professionals at attacking or securing systems like the criminal or government hackers tasked with uncloaking them. The attacks on Tor, by NSA & by academics, apply where many users get unmasked remotely in a way they never see. They can all be surveilled at once by NSA's existing infrastructure. Outside this, the FBI deanonymized users by using a browser 0-day (endpoint attack). Five Eyes, esp NSA, have 0-days on each desktop and mobile OS with high, market share per leaks. We regularly see Russian and Chinese hackers breach them in the news. We also see all kinds of hackers breach systems in general plus lots of bad decisions by users. The risk of a stealthy, mass, inexpensive leak of people using digital communications is very high whereas the odds of them outthinking average surveillance is low and targeted hacking very low. So, let's look at people using meatspace with methods like I described above and simple guides on their use. They've spent their whole life learning how to tell when people are watching them, nervous, or just acting unusual. They have numerous ways of blending into a crowd where WiFi or drops are available but with limited camera coverage. They can learn the traffic patterns of the area to see if anyone sticks out. Likewise, there's private spots to stash a long-range antenna and camera to do the same from a distance. Many know how to ditch evidence or pay for stuff with cash. Keeping their name and I.P. out of it reduces odds they'll be seen where no tech or HTTPS-style comms increases odds surveillance ignores them entirely as mere Internet, background noise. These methods were and are used successfully by many crooks and spies... today just like decades ago with drops, radios, and countersurveillance... while "smart" people, their OS's, and anonymity software get owned on computers quite regularly. So, the average person looks to be better off avoiding the Internet, executable media, or anything that draws attention when trying concealed communication with other average people. It's easier for them to follow this advice like that than learn arcane, technical information that changes constantly. It also turns inexpensive, stealthy, mass attacks into requiring footsoliders focused on one person or a small group at a time risking detection constantly. That's also with no guarantee they'll keep whatever gains they acquire in intelligence collection. That's a significant increase in effectiveness versus new methods where you trust things that can be and are often invisibly hacked by remote attackers. "When I go out, I can never be confident I'm not being followed, recorded in various ways or leaking biometric evidence. Online, I can operate with an assumption of full compromise _and still come up with useful OPSEC strategies_." No you can't. If you can't trust or mitigate physical risks, then you can be compromised in the physical realm to defeat your software countermeasures. There's whole Snowden slides dedicated to those things. Not to mention books by experts in hacking, black bag jobs (esp TSCM), infiltration (esp spec ops), and crime (esp B&E's or cons). Physical security and lack of subversion is a prerequisite for and/or major part of online security. So, what you're really doing is hoping you're not worth their time or money at all plus no flaws are found in your hardware, firmware, OS, browser, Tor, or any other running software. Might be true but lots of counterevidence on software side. ;)