Slashdot Mirror
Swedish Researchers Expose China's Tor-Blocking Tricks
An anonymous reader writes "A pair of researchers at Karlstad University have been able to establish how the Great Firewall of China sets about blocking unpublished Tor bridges. The GFC inspects web traffic looking for potential bridges and then attempts 'to speak Tor' to the hosts. If they reply, they're deemed to be Tor bridges and blocked. While this looks like another example of the cat and mouse game between those wishing to surf the net anonymously and a government intent on curtailing online freedoms, the researchers suggest ways that the latest blocking techniques may be defeated."
I hatelove slashdot summaries, and here is another example of why. Yes, I haven't read TFA.
When you use the word "while" like this, it sounds like you're going to be contradicting the first point. Especially when you use the phrase "this looks like" immediately afterward.
Instead, the second part of the sentence goes on to directly corroborate what the scenario looks like. Surprise! While it looks like you're setting up a contradiction, you finish up with reinforcement.
So in fact it doesn't just "[look] like another example of the cat and mouse game", but in fact it literally is an example of the cat and mouse game, and the researches propose another way for the mouse to escape. And yet another awkward summary graces the Slashdot homepage, in the grand tradition.
The fundamental problem here is that Tor is accessible to the public. No, you read that right. As long as anyone can download a Tor client and connect, that person will have the IP address of at least one other Tor user. There is very little that can be done to prevent this without limiting access to the Tor network by some kind of handshake/authentication model. At the very least, the network is vulnerable to a denial of service attack; Since it can't tell a legitimate user from an illegitimate one: By design, the traffic is encrypted and the source obfusciated.
Tor can't ever fully succeed in its objective -- it can only maintain network integrity so long as the ratios between different types of users, client accesses, etc., remain in the green. Should the balance ever tilt, the network will become unusable.
A real solution is end to end encryption network-wide, which is what IPv6 was supposed to do, but as I'm sure you've all realized; the capitalist owners of the routers, switches, ISPs, etc., have decided artificial scarcity of IP address space could be profitable to them, so IPv6 is sort of dead on arrival. But even if it weren't, the notion that the ISP can't control what connections are made based on content is not something any of them want to give up; again, in the name of profits.
So basically, we need a whole new internet, built by the people, from the ground up. And it will probably have to be wireless. The problems of wireless high speed internet between buildings is hard enough; Try between cities. :\ But that's the only way I see of re-establishing a free and democratic digital communications medium.
#fuckbeta #iamslashdot #dicemustdie
The throughput is reasonable, but the latency is pretty high.
In China people are seeking different ways to breach GFW. We mainly use SSH tunnel, OpenVPN, or some sorts of HTTPS proxy (with some obfuscation needed by both sides or it doesn't work for GFW has capacity to probe SSL/TLS proxy).
I am glad tor now is functioning again in China. Just began to spread the obfsorxy tor browser to the others who need it.