4 ms·

It seems that there are at most 10000 Tor nodes. Are the identity of the node operators known? Do known trusted organizations and individuals own the majority

by aborsy 3y ago

It seems that there are at most 10000 Tor nodes. Are the identity of the node operators known? Do known trusted organizations and individuals own the majority of these nodes? It’s indeed expensive and risky for individuals to run Tor notes.

sdsd 3y ago
I know that Chaos Computer Club used to run a bunch of them. And Noisebridge did for a while, but I think they stopped. A few universities, too. I haven't been very active in the space for about a decade, I'd also love a more knowledgeable answer
dpifke 3y ago
See https://nusenu.github.io/OrNetStats/ https://nusenu.github.io/OrNetStats/ It's up to individual relay operators whether or not to publish contact information. For an example, see one of my relays here: https://nusenu.github.io/OrNetStats/w/relay/375DCBB2DBD94E5263BC0C015F0C9E756669617E.html https://nusenu.github.io/OrNetStats/w/relay/375DCBB2DBD94E52... (If you're not an exit relay, it's neither risky nor expensive.)
127361 3y ago
Try the I2P anonymous network where by default every user is also a router. You can chain the two together, set I2P to use an outproxy and set the Tor browser to connect through I2P. I have no clue as to what security this provides nowadays. I predict the NSA or FBI have large scale packet timing correlation in operation now. Or the random number generator has been compromised, or there are a series of bugs in the Tor or I2P implementation itself? I also think there was a human compromise of the Tor Project itself 8-9 years ago as well, it has been significantly weakened. Personally I tend to like one-way data broadcasting systems, that way the receivers cannot be traced if they are air-gapped, e.g. satellite or radio data broadcasting. However nobody operates any useful service nowadays. Twenty years ago you could receive the whole worldwide Usenet feed with a DVB-S PCI card, unencrypted in the clear from a service called Cidera[1]. It was just UDP multicast packets containing Usenet messages, split into fragments with sequence numbers, trivial to reverse engineer. I had this at service home, we had dial-up and with a 55cm satellite dish in my bedroom window, I had a 45Mb/s data feed. I wrote my own software including a device driver for the DVB-S card running under FreeBSD (4.2-RELEASE I believe). Nearly 1000x faster than dial up, which was mind-blowing back then. Also archiving Nostr text messages is interesting, you can get the entire worldwide feed of messages from the relays, and then search for whatever topic of interest you want in the messages. Nobody can tell what you are searching for, unless your computer is compromised. I'm thinking about broadcasting these by satellite somehow, but satellite bandwidth is very expensive. It's just a matter of getting the funds for it. Reception should be possible using a RTL-SDR or AD9364+USB3.0 microcontroller[2], and a satellite TV dish. The AD9364 chip can be had from AliExpress[3] for $6 now. 1. https://web.archive.org/web/20020806064624/http://www.cidera.com/services/usenet_news/index.php https://web.archive.org/web/20020806064624/http://www.cidera... 2. https://www.crowdsupply.com/amungo-navigation/stixrf https://www.crowdsupply.com/amungo-navigation/stixrf 3. https://www.aliexpress.com/i/33017832196.html https://www.aliexpress.com/i/33017832196.html
- mdip 3y ago
  I thought I was a veteran of the early internet/Usenet enough to know most of the ins and outs but this DVB-S Usenet is new to me. That sounds like a pretty unique approach to Usenet content delivery. I wonder if anything else/similar has existed/exists (unlikely in Usenet and I can't think of something quite as useful).
rendx 3y ago
The goal is to not have to trust the node operators, but to have as much diversity as possible. They can only deanonymize you if all three nodes in one cascade collude. Running a non-exit node carries very little to no risk.
- throwaway89201 3y ago
  No, you only need to compromise the first node – named the entry guard for that reason [1] – and either the exit node or ideally the endpoint (hidden) service. Deanonymization is then possible by correlating the timing of traffic between those two points, as Tor wants to be low-latency, without randomly delaying traffic. For this reason not all nodes may be guard nodes, as decided by the directory authorities, and guard nodes are maintained for a longer time by the client to reduce the chance that you pick a compromised guard node because you switch often. This is balanced against the risk that you are unlucky and pick a compromised guard at first (which you then maintain for a longer time). The exit node is pretty much assumed to be compromised, as it's a role not available to many entities – it requires high bandwidth and much teeth-gritting – and the public internet is intercepted at large anyway. [1] https://support.torproject.org/about/entry-guards/ https://support.torproject.org/about/entry-guards/
joe_the_user 3y ago
The reason Tor exists is because of intelligence agencies - they literally created it[1]. It makes perfect sense. Agencies certainly want to be able to surveil the population at large but they want to be able to act anonymously themselves. I mean, any fixed entry-point an agency chooses to the Internet is going to get a lot of attention. Agencies could mix it up in various ways - change entry points, mix their traffic, etc ... and the final result looks a lot like Tor. And so yeah, I'm sure you have idealistic operators who take heat for running nodes but I'd expect the agencies put some resources into making sure one way or another that the exit nodes exist. [1] https://en.wikipedia.org/wiki/Tor_(network)#History https://en.wikipedia.org/wiki/Tor_(network)#History
forward1 3y ago
Given the bandwidth cost alone of some of the top nodes, it is not unreasonable to draw conclusions about who funds and operates them. With the additional incentive to control the network considered, the truth of the matter couldn't be more obvious.

by aborsy 3y ago

sdsd 3y ago

I know that Chaos Computer Club used to run a bunch of them. And Noisebridge did for a while, but I think they stopped. A few universities, too. I haven't been very active in the space for about a decade, I'd also love a more knowledgeable answer

dpifke 3y ago

See https://nusenu.github.io/OrNetStats/ https://nusenu.github.io/OrNetStats/ It's up to individual relay operators whether or not to publish contact information. For an example, see one of my relays here: https://nusenu.github.io/OrNetStats/w/relay/375DCBB2DBD94E5263BC0C015F0C9E756669617E.html https://nusenu.github.io/OrNetStats/w/relay/375DCBB2DBD94E52... (If you're not an exit relay, it's neither risky nor expensive.)

127361 3y ago

Try the I2P anonymous network where by default every user is also a router. You can chain the two together, set I2P to use an outproxy and set the Tor browser to connect through I2P. I have no clue as to what security this provides nowadays. I predict the NSA or FBI have large scale packet timing correlation in operation now. Or the random number generator has been compromised, or there are a series of bugs in the Tor or I2P implementation itself? I also think there was a human compromise of the Tor Project itself 8-9 years ago as well, it has been significantly weakened. Personally I tend to like one-way data broadcasting systems, that way the receivers cannot be traced if they are air-gapped, e.g. satellite or radio data broadcasting. However nobody operates any useful service nowadays. Twenty years ago you could receive the whole worldwide Usenet feed with a DVB-S PCI card, unencrypted in the clear from a service called Cidera[1]. It was just UDP multicast packets containing Usenet messages, split into fragments with sequence numbers, trivial to reverse engineer. I had this at service home, we had dial-up and with a 55cm satellite dish in my bedroom window, I had a 45Mb/s data feed. I wrote my own software including a device driver for the DVB-S card running under FreeBSD (4.2-RELEASE I believe). Nearly 1000x faster than dial up, which was mind-blowing back then. Also archiving Nostr text messages is interesting, you can get the entire worldwide feed of messages from the relays, and then search for whatever topic of interest you want in the messages. Nobody can tell what you are searching for, unless your computer is compromised. I'm thinking about broadcasting these by satellite somehow, but satellite bandwidth is very expensive. It's just a matter of getting the funds for it. Reception should be possible using a RTL-SDR or AD9364+USB3.0 microcontroller[2], and a satellite TV dish. The AD9364 chip can be had from AliExpress[3] for $6 now. 1. https://web.archive.org/web/20020806064624/http://www.cidera.com/services/usenet_news/index.php https://web.archive.org/web/20020806064624/http://www.cidera... 2. https://www.crowdsupply.com/amungo-navigation/stixrf https://www.crowdsupply.com/amungo-navigation/stixrf 3. https://www.aliexpress.com/i/33017832196.html https://www.aliexpress.com/i/33017832196.html

mdip 3y ago
I thought I was a veteran of the early internet/Usenet enough to know most of the ins and outs but this DVB-S Usenet is new to me. That sounds like a pretty unique approach to Usenet content delivery. I wonder if anything else/similar has existed/exists (unlikely in Usenet and I can't think of something quite as useful).

rendx 3y ago

The goal is to not have to trust the node operators, but to have as much diversity as possible. They can only deanonymize you if all three nodes in one cascade collude. Running a non-exit node carries very little to no risk.

throwaway89201 3y ago
No, you only need to compromise the first node – named the entry guard for that reason [1] – and either the exit node or ideally the endpoint (hidden) service. Deanonymization is then possible by correlating the timing of traffic between those two points, as Tor wants to be low-latency, without randomly delaying traffic. For this reason not all nodes may be guard nodes, as decided by the directory authorities, and guard nodes are maintained for a longer time by the client to reduce the chance that you pick a compromised guard node because you switch often. This is balanced against the risk that you are unlucky and pick a compromised guard at first (which you then maintain for a longer time). The exit node is pretty much assumed to be compromised, as it's a role not available to many entities – it requires high bandwidth and much teeth-gritting – and the public internet is intercepted at large anyway. [1] https://support.torproject.org/about/entry-guards/ https://support.torproject.org/about/entry-guards/

joe_the_user 3y ago

The reason Tor exists is because of intelligence agencies - they literally created it[1]. It makes perfect sense. Agencies certainly want to be able to surveil the population at large but they want to be able to act anonymously themselves. I mean, any fixed entry-point an agency chooses to the Internet is going to get a lot of attention. Agencies could mix it up in various ways - change entry points, mix their traffic, etc ... and the final result looks a lot like Tor. And so yeah, I'm sure you have idealistic operators who take heat for running nodes but I'd expect the agencies put some resources into making sure one way or another that the exit nodes exist. [1] https://en.wikipedia.org/wiki/Tor_(network)#History https://en.wikipedia.org/wiki/Tor_(network)#History

forward1 3y ago

Given the bandwidth cost alone of some of the top nodes, it is not unreasonable to draw conclusions about who funds and operates them. With the additional incentive to control the network considered, the truth of the matter couldn't be more obvious.