5 ms·
Tor is not, nor has it ever been, trustworthy. Hell, you can still try active deanonymization for youself: https://github.com/Miserlou/Detour https://github.com
by Mizza 10y ago
Tor is not, nor has it ever been, trustworthy. Hell, you can still try active deanonymization for youself: https://github.com/Miserlou/Detour https://github.com/Miserlou/Detour
This didn't used to be a problem, as it was essentially run as a sandbox project for the academic anonymity community. It was very up front about its capabilities and limitations.
Unfortunately, in recent years, the US government has been bankrolling more "privacy" software development through its propaganda arms (OTF, RFA, etc.), and the Snowden revelations have led private foundations to follow suit.
As such, the organization doubled down on rebranding to be a "human rights" _tool_, as this is what grant giving organizations love to promote (free speech in Iran, activist publishing, etc.) This combined with a overly-enthusiastic do-gooders gaining more and more prominence in the Tor organization has led to the dangerous situation of promoting inherently insecure software as a security solution to vulnerable people. This is a general problem in the scene (remember when those activists in South America got vanned for using CryptoCat?) - and one that I've been guilty of myself in the past.
I really hope the new boards steers them back to the academic realm and slaps a big red USE AT YOUR OWN RISK warning on the tin. Unfortunately, I think the opposite will happen.
- PieterH 10y agoA more dangerous side-effect of branding Tor as a "human rights" tool is that whereas if it's a "privacy" tool, people in oppressive regimes can legitimately use it (e.g. people working for such regimes), whereas as a "human rights" tool, its simple presence on a computer is evidence of guilt.
- bruo 10y agoI haven't live in a lot of oppressive regimes, but usually this is not how it works. I lived two close examples When i was a kid a friend of my grand mother was arrested because a neighbor said he was a communist. He was tortured for a week and his party, kind of center/right wing took him out as they were a close party to the current government. Oppressive regimes usually don't need evidences. A friend of mine was arrested for two years, accused for terrorism. The proofs? a war and peace copy (not even a photocopied book) and a guns and roses poster. And this was in "democracy"... so stupid proofs are also used, and whatever can be a proof, like a book about cubism was considered that was a book of cuba's ideology. Tor has been for years looked by "regular"/"normal"/"common" people as a tool for drug dealers or child molesters. The switch to a human rights tool doesn't seem to really put it more into the illegal line. Anyway, oppressive regimes do whatever they want, Tor can avoid some of the spying but if the state is already taking your computer you are screw up with Tor or without it.
- nickpsecurity 10y agoI think you're missing his point. Many people in oppressive regimes, including oppressors, are interested in a bit of extra privacy. Regime might even be able to get around it since a lot of "privacy" tools are bullshit. A tool saying it's specifically designed to defeat oppressive regimes is practically an attack on them. Even possessing it says you're fighting the regime. So, eliminating all such tools or people using them is a logical response for oppressive regimes. And people on Pieter's side of this issue probably think it also logically follows to prevent that mental connection from happening in government officials' minds. Just gotta change how it's branded.
- bruo 10y agoI understand the point, what I am saying is that things don't work this way, neither in my personal experience (which i told) and neither in the areas i work (i work for a human rights organization, which doesn't recommend any software btw). There is a difference on what you think people should react and what people really react. If it's as concepts, "human rights" triggers less alarms than anonymity, first because while I don't know which regimes you consider oppressive there is a big probability they themselves think they comply and/or promote human rights. Iran for example have a Islamic Human Rights Commission and proudly promote it. Israel would be another country that fits this example. Then we have that anonymity could mean something it scares them the most, which is not human rights defenders but spying. Tor is already in a bad list for this reason, same as any anonymity software. The biggest threat those countries face is still military intervention or terrorism. A friend was arrested while taking pictures in Palestine, when questioned he was asked if he had Tor or i2p installed, PGP or any encryption software on his laptop. They didn't took his laptop away, but that was before the switch to "human rights" brand. Then there is another vector we can take, Tor as circunvention. Another friend when visiting sudan got a pamflet to not use Tor, VPNs or Proxies when asked for the visa. The hotel made the same requeriment. This was 4 years ago. The reason was not that Sudan has been in the list of the worst human rights offenders but that you could access immoral content with it. So, while I understand the point, it doesn't seem to have a backed reality to be sustained.
- wybiral 10y agoMost of the issues seem to revolve around accessing clearnet resources through Tor. Hidden services within Tor still seem secure (whatever that word means these days).
- kevinr 10y ago> This combined with a overly-enthusiastic do-gooders gaining more and more prominence in the Tor organization has led to the dangerous situation of promoting inherently insecure software as a security solution to vulnerable people. What is the inherently secure alternative available to these vulnerable people? > I really hope the new boards steers them back to the academic realm and slaps a big red USE AT YOUR OWN RISK warning on the tin. What causes you to believe that activists and vulnerable people would stop using Tor if this warning were in place?
- Mizza 10y agoAre you asking what they should use instead? Because there is no alternative. My point isn't that they should be using some other specific piece of software, or even that they shouldn't be using Tor, just that they should be aware that Tor is old software with many known vulnerabilities, a poor track record of security including active RCE attacks by governments and private companies, etc. If you go to the homepage, you'll see none of that. This might not deter people from using Tor entirely, but at least they'd understand the threat model a bit better so they could change their behavior accordingly, for instance maybe not using it at their own home, not using it for Facebook, etc. As much as we might want to believe that we can code our way out of anything, human rights demands more than software.
- jesseb 10y ago> Because there is no alternative. Just out of curiosity and for purely educational purposes (I admittedly know very little about the topic), how about something like I2P or FreeNet? Are these at all viable alternatives? What other choices do we have for "anonymity" outside of the usual VPN/Proxy?
- Mizza 10y agoAnonymity isn't binary. Perfect anonymity is, and always will be, completely impossible. Practically speaking, we can only talk about reducing risk in different scenarios. I like I2P as it has a reduced threat footprint compared to Tor as it's all internal, but it's also a smaller network, probably more vulnerable to Sybil, less audited code, etc. Again, my point is that there is no perfect solution, it all depends on the situation and the requirements.
- nickbail3y 10y agoI'm not sure where this narrative is coming from. TOR was developed as an anonymous network by the US Naval Research Lab. It was designed for use by military and intelligence. TOR was never just some academic experiment. TOR is still a valid tool. No, it wasn't designed to foil NSA level surveillance, because it was built by the US. But this vulnerability isn't even related to TOR, it has to do with the TOR Browser. The Snowden leaks contain slides where the NSA clearly laments the use of TOR, so saying that it never has been trustworthy is simply not true.
- Mizza 10y agoIt was "designed" for military use in the sense they were computer scientists working for the Naval research, but the community felt like an academic one. For instance, it was originally published at USENIX. The majority of discussion of Tor was related to papers on Anonbib, not about code itself. TBB didn't even exist until much later. I've never seen any evidence that it was ever used "in production" by the military before it was made public. Re: The NSA Tor slides - they're really not as damning as you say - http://i.imgur.com/cnOeVQf.png http://i.imgur.com/cnOeVQf.png - and they're also made before the FBI was caught using remote code execution exploits against Tor users.
- nickbail3y 10y ago> but the community felt like an academic one How do you know what the development environment 'felt' like? Are you Roger Dingledine or Nick Mathewson? And I doubt that you will find any evidence that it was used before being made public. Using TOR before it was public would be like screaming, "HEY I'M HIDING SOMETHING! AND I'M US MILITARY OR INTELLIGENCE!". The whole point of releasing it was to gather a userbase. Otherwise, TOR wouldn't be very anonymous at all.
- Mizza 10y agoI was working on a related anonymity project at the same time, funded by the same grant giving organization. I lurked and participated, I have met RD and NM IRL, and I have been a hidden service operator for many years. That being said, I am only at the far periphery of the project. But, I have been out here for a while and I stand by my original statement.
- openasocket 10y ago> Tor is not, nor has it ever been, trustworthy. Hell, you can still try active deanonymization for youself: https://github.com/Miserlou/Detour https://github.com/Miserlou/Detour Is it even possible to protect against end-to-end time correlation attacks without massively increasing latency?
- Mizza 10y agoI'd call that an "open problem."
- openasocket 10y agoI'm fairly certain it isn't possible. If you can detect, with high confidence, that an entity received a message and that an entity sent a message, timing attacks can be done. The traditional means to prevent this involves sending dummy traffic, introducing unpredictable latency, or using some sort of broadcast to conceal the exact receiver. None of those things are practical for a low-latency proxy. It's been a couple years since I studied this, so I might be wrong.