Tor: The Second-Generation Onion Router (2004)(svn.torproject.org)
svn.torproject.org
Tor: The Second-Generation Onion Router (2004)
https://svn.torproject.org/svn/projects/design-paper/tor-design.html
71 comments
I think it's an important distinction that one has identified developers, and one has anonymous developers.
It doesn't seem clear to me that one is generally better than another, I think they complement each other.
It doesn't seem clear to me that one is generally better than another, I think they complement each other.
It mentions mixmaster and mixminion, but those projects seem to have died. I feel like they are needed more than ever.
Indeed. I've often thought that such high-latency mix-nets (with their potentially superior resistance to traffic analysis) would be a nice complement to low-latency networks like Tor.
Have a look at Katzenpost. They are trying to build a modern, high-latency, mix network.
https://katzenpost.mixnetworks.org/
https://katzenpost.mixnetworks.org/
I2P provides cover traffic. I believe GNUnet can also be configured to use a certain amount of cover traffic.
The downside is that browsing the web on a service like that sucks ass. Tor trades off some security for a basic level of user experience.
Sadly, modern web design makes stuff like WWWoffle less useful than it used to be. Plus it's difficult to set up fully anonymous services on the far side of the link, I mean you're using Tor for a reason, and leaving breadcrumbs on the far side of the network is risky.
Sadly, modern web design makes stuff like WWWoffle less useful than it used to be. Plus it's difficult to set up fully anonymous services on the far side of the link, I mean you're using Tor for a reason, and leaving breadcrumbs on the far side of the network is risky.
Using TBB has gotten much better over the past few years, often you can stream YouTube without much issue.
Tor is great, but it's always been a bit slow. Now that cryptocurrency is a thing, I often wonder if that problem could be solved by incentivizing the running of relays and exit nodes through microtransactions without compromising privacy. Would be great if we could make Tor fast enough that it could actually be used as a substitute for commercial VPNs.
That reminds me of this little onion routering thingy i made that spiraled out of control into a semi functioning project that I got hired to work on full time
(https://github.com/loki-project/loki-network)
(https://github.com/loki-project/loki-network)
Also a dev on loki-network, we have $4.1+ million USD so far invested in relays (and potentially exits). It's layer 3 (so think more like i2p) but newer crypto and much better latency.
A good overview of the project is here: https://github.com/loki-project/loki-network/blob/master/doc...
other docs here: https://github.com/loki-project/loki-network/blob/master/doc...
Join our test-net now: https://discord.gg/eB8k6xQ or #llarp on freenode
A good overview of the project is here: https://github.com/loki-project/loki-network/blob/master/doc...
other docs here: https://github.com/loki-project/loki-network/blob/master/doc...
Join our test-net now: https://discord.gg/eB8k6xQ or #llarp on freenode
It's not too slow these days. I routinely browse over Tor, including watching YouTube videos, and it doesn't feel substantially different to just my ordinary ADSL connection.
That is being worked on! It’s called Orchid:
https://www.orchid.com/
https://www.orchid.com/
Is there more info about active development? That site doesn't say much. I found some GitHub repos that were last updated in the summer[0]. It's going to be unfortunate if they stay with AGPL.
0 - https://github.com/orchidsource
0 - https://github.com/orchidsource
(I am now in charge of technology at Orchid. None of that source code is relevant to our project. I am sorry for the confusion.)
Can you explain why you are concerned with the usage of AGPL? Are you hoping to make a closed-source hosted fork eventually?
Can you explain why you are concerned with the usage of AGPL? Are you hoping to make a closed-source hosted fork eventually?
> Can you explain why you are concerned with the usage of AGPL? Are you hoping to make a closed-source hosted fork eventually?
This can get very copy-left philosophical very quickly, so I'll be brief. For transit technology, especially if adoption matters, the less a license infects downstream projects the better for ecosystem growth. For example, I have an open source Tor library that is MIT licensed and it statically links Tor. I want the world to use my library, commercial/closed and open alike. I would never be able to do that with Orchid and therefore would not touch the tech. Make what you build on your network AGPL, but the underpinnings themselves should be as restrictionless as possible. These are my opinions and I understand the other opinion of "if you want the right to use our stuff, you have to abide by our openness rules on your stuff" but just know that you are trading adoption for principles.
This can get very copy-left philosophical very quickly, so I'll be brief. For transit technology, especially if adoption matters, the less a license infects downstream projects the better for ecosystem growth. For example, I have an open source Tor library that is MIT licensed and it statically links Tor. I want the world to use my library, commercial/closed and open alike. I would never be able to do that with Orchid and therefore would not touch the tech. Make what you build on your network AGPL, but the underpinnings themselves should be as restrictionless as possible. These are my opinions and I understand the other opinion of "if you want the right to use our stuff, you have to abide by our openness rules on your stuff" but just know that you are trading adoption for principles.
FWIW, you are in fact talking to a team of people who are core Free Software people (such as Brian Fox, who was the first person employed by the Free Software Foundation)... but like, it must be pointed out that your Tor library is almost certainly a client library, and the license of the Tor server you are accessing would not affect the ability for people to use it: "statically linking Tor" is kind of simplistic, as it assumes Tor is a monolithic component (and maybe it strangely is... I honestly have never looked much at its code).
In case it helps immediately clarify: you absolutely will be able to use the Orchid platform to build commercial, closed source distributed, semi-distributed, and even centralized applications.
For philosophical background on this, I encourage reading the position of Richard Stallman on why the Ogg Vorbis library should not be licensed under GPL. Our mission (to decentralize software and communication) causes similar quirks.
https://lwn.net/2001/0301/a/rms-ov-license.php3
(Exactly what the boundaries are on the components and under what license they will be released is thereby not yet finalized as we continue to work on architecture, but there are at least a couple things we are building that will seriously be released under 0-BSD.)
In case it helps immediately clarify: you absolutely will be able to use the Orchid platform to build commercial, closed source distributed, semi-distributed, and even centralized applications.
For philosophical background on this, I encourage reading the position of Richard Stallman on why the Ogg Vorbis library should not be licensed under GPL. Our mission (to decentralize software and communication) causes similar quirks.
https://lwn.net/2001/0301/a/rms-ov-license.php3
(Exactly what the boundaries are on the components and under what license they will be released is thereby not yet finalized as we continue to work on architecture, but there are at least a couple things we are building that will seriously be released under 0-BSD.)
I appreciate your efforts and all efforts to improve privacy. The library I was referencing is actually for building Tor servers too, though even just its client part communicates with either a local Tor daemon or the statically linked Tor in the executable, both of which would be subject to the AGPL regardless of the server part. But if the principles are similar to the Ogg lib and liborchid is LGPL/BSD/MIT/Apache/etc then we are in complete agreement. Consider me as one of those that, when the core lib is built/available, will make things with it.
Hey j, we did work together in another life. I met brian fox at a conference in Boston earlier this year that he spoke at, but haven’t heard anything about Orchid since. Shoot me a note on Twitter in my profile if we could find time to chat.
Just want to chime in and say thank you for building free software. :)
In my experience, Tor runs about as well as ProtonVPN with their "Secure Core" option enabled (which adds routing through an additional hop located somewhere in the EU). Tor has pretty good throughput, I think the problem is the the additional latency that comes from routing through multiple nodes.
It has been getting faster lately, and I mean last couple of years. Or maybe my tolerance grew from using LTE but I'm pretty sure it has been getting faster.
Tor's slowness (often overstated) is more about hop count and geographic diversity than relay/exit node count.
Tor is great! Sadly HN shadow bans all accounts created with it. Lots of other services also won’t let you sign up or require a phone verification if you use it.
Apparently not safe for work
A company could have a lot of problems if one of their employees decided to host a node in their network, so quite literally yeah
What kind of problems company could have?
I ran an exit node at home for a couple months. Within minutes of my dynamic IP being updated from my ISP, cloudflare could figure out that I was an exit node, and then would fail to differentiate my own, non tor traffic, with that of the traffic exiting my node. The result being every single cloudflare backed website will hit you with a captcha on every page you view.
My ISP would also randomly shut off my connection with some vague email about having compromised computer on my network i needed to fix. I'd have to call and make up some story to get them to turn my internet back on.
Essentially all kinds of people associate TOR with unsavory activity and as such put hoops to jump through when accessing stuff through tor. When you host an exit node your non tor traffic is considered to be outbound TOR traffic based on the IP the traffic originates from.
I cant imagine what would happen if I ran a node on the same network as our email server. I have to assume that we'd get blacklisted for spam quick if our mx resolved to a tor exit node IP.
Then there's the whole issue of more or less being guaranteed to be complicit in the distribution of CP. With really only safe harbor laws protecting you.
My ISP would also randomly shut off my connection with some vague email about having compromised computer on my network i needed to fix. I'd have to call and make up some story to get them to turn my internet back on.
Essentially all kinds of people associate TOR with unsavory activity and as such put hoops to jump through when accessing stuff through tor. When you host an exit node your non tor traffic is considered to be outbound TOR traffic based on the IP the traffic originates from.
I cant imagine what would happen if I ran a node on the same network as our email server. I have to assume that we'd get blacklisted for spam quick if our mx resolved to a tor exit node IP.
Then there's the whole issue of more or less being guaranteed to be complicit in the distribution of CP. With really only safe harbor laws protecting you.
People associate TOR with unsavory activity because TOR is used for unsavory activity. Running an exit node on a link you care about is insanity.
The main reason I have never run an exit node is fear that some overzealous LEO trying to make a name for himself is going to trace an IP off of a kiddie porn or drug market to my house and send in SWAT to break all of my windows and tie me up in court for months before deciding that charging me as an accomplice probably won't work this time because I didn't opt for the public defender.
The main reason I have never run an exit node is fear that some overzealous LEO trying to make a name for himself is going to trace an IP off of a kiddie porn or drug market to my house and send in SWAT to break all of my windows and tie me up in court for months before deciding that charging me as an accomplice probably won't work this time because I didn't opt for the public defender.
Exit node, sure, but hosting a regular node should not pose any threat.
My impression is that TOR is exit node-constrained. Is this correct?
If you're using onion-site, exit nodes are not used at all, so it depends on your use-case. For accessing internet probably yes, exit-nodes are limiting factor.
News?
I keep copies of these papers in a safe in case I never need to re-develop the Internet from scratch. No seriously, the Tor papers are amazing and even hint at the means by which Tor might be manipulated to, say, track down someone like Ross Ulbricht, founder of the Silk Road, who is currently serving a double life sentence, plus forty years, without the possibility of parole.
I myself have read this particular paper a few times, and plan to put a fresh printed copy in a cover protector. Something about keeping this thing for years strikes a chord with me
>Ross Ulbricht, founder of the Silk Road, who is currently serving
For running a free market. It's fucked up.
For running a free market. It's fucked up.
Stop trying to rewrite history. He was convicted of running the market, but the reason the judge threw the book at him is because he tried to have five people killed.
>Because it is contested, the Court must make appropriate factual findings if it is to include it. The standard by which I do that is by a preponderance of the evidence. Ulbricht's directed violence here is and relates to the murders for hire which he is alleged to have commissioned and paid for. The Court must determine whether these allegations have been demonstrated by a preponderance of the evidence and I find that there is ample and unambiguous evidence that Ulbricht commissioned five murders as part of his efforts to protect his criminal enterprise and that he paid for these murders. There is no evidence that he was role-playing.
https://freeross.org/wp-content/uploads/2015/05/Sentencing_2...
Maybe you think he didn't try to commission murders. Maybe you think judges shouldn't have so much latitude in deciding sentences. But the simple fact remains: Ulbricht is serving two life sentences mostly because the judge found that he tried to kill five people.
>Because it is contested, the Court must make appropriate factual findings if it is to include it. The standard by which I do that is by a preponderance of the evidence. Ulbricht's directed violence here is and relates to the murders for hire which he is alleged to have commissioned and paid for. The Court must determine whether these allegations have been demonstrated by a preponderance of the evidence and I find that there is ample and unambiguous evidence that Ulbricht commissioned five murders as part of his efforts to protect his criminal enterprise and that he paid for these murders. There is no evidence that he was role-playing.
https://freeross.org/wp-content/uploads/2015/05/Sentencing_2...
Maybe you think he didn't try to commission murders. Maybe you think judges shouldn't have so much latitude in deciding sentences. But the simple fact remains: Ulbricht is serving two life sentences mostly because the judge found that he tried to kill five people.
> Maybe you think judges shouldn't have so much latitude in deciding sentences.
Sure scares me. Why wasn't he tried for the other crimes if his punishment is going to be based on them? This isn't a little difference to take into account when sentencing, these are important crimes on their own deserving a right to trial.
Sure scares me. Why wasn't he tried for the other crimes if his punishment is going to be based on them? This isn't a little difference to take into account when sentencing, these are important crimes on their own deserving a right to trial.
Because you take them down how you can; in a similar light how Capone was taken down on taxes and not murder, corruption, smuggling, or conspiracy.
Or are all purveyors of the 'free market' automatically heroes in your eyes?
Or are all purveyors of the 'free market' automatically heroes in your eyes?
> Or are all purveyors of the 'free market' automatically heroes in your eyes?
No, I don't even like the guy. Why must I consider someone a hero to feel they deserve a trial for their punished actions?
No, I don't even like the guy. Why must I consider someone a hero to feel they deserve a trial for their punished actions?
A trial did occur for Ross Ulbricht; I don't understand your point.
If someone is accused of X, but you "get them" for Y, and the sentencing is based on the fact that they "did" X... you never had a trial that proved that they did X, so anything you think you know about their guilt of X shouldn't be relevant to their sentence for Y. Because you don't really know.
And, even if you know mathematically without a doubt that "they did it" in point-of-fact, and you have all the evidence to back you up such that you're really sure that if you did have a trial, it'd be an open-and-shut case... you still can't assume. Because it's not about whether you see guilt or innocence when you look at the evidence; it's about whether the jury sees guilt or innocence when they look at the evidence.
Jury nullification means a jury can just decide, arbitrarily, that somebody's not guilty of something. And then double-jeopardy means that you can't ask that question again. It's been resolved, permanently: the accused has been declared innocent, in the eyes of the law—however strong your proof was! And any further judgements by the courts have to take, as input, that innocent verdict that came out of the trial; not the proof-of-guilt that went in but didn't survive.
What that implies, to me, is that in the Al Capone example, he should never have been treated as anything other than a regular tax evader, and should only have received a regular tax-evader's sentence.
I'm reminded heavily of the recent p-hacking controversy in science. Imagine a world where we "pre-register" a trial when an investigation is begun, and are then forced to go through with it (and pretend the government has the infinite money required to enable this.) In this world, that trial would likely declare the suspect innocent a lot of the time... and all the evidence gathered to do so would be "used up" by that trial. You couldn't turn around and use the same evidence to prove that "because he's guilty of X, it's more likely that..." anything else. Because, in the eyes of the law, he isn't guilty of X, and anything that proves that, doesn't. The evidence itself was nullified when the verdict was.
In the real world, just to save time and money, we don't bother to prosecute trials we know we'll lose. But shouldn't the effects still be the same as if we did? An optimization shouldn't change the semantics of the system.
And, even if you know mathematically without a doubt that "they did it" in point-of-fact, and you have all the evidence to back you up such that you're really sure that if you did have a trial, it'd be an open-and-shut case... you still can't assume. Because it's not about whether you see guilt or innocence when you look at the evidence; it's about whether the jury sees guilt or innocence when they look at the evidence.
Jury nullification means a jury can just decide, arbitrarily, that somebody's not guilty of something. And then double-jeopardy means that you can't ask that question again. It's been resolved, permanently: the accused has been declared innocent, in the eyes of the law—however strong your proof was! And any further judgements by the courts have to take, as input, that innocent verdict that came out of the trial; not the proof-of-guilt that went in but didn't survive.
What that implies, to me, is that in the Al Capone example, he should never have been treated as anything other than a regular tax evader, and should only have received a regular tax-evader's sentence.
I'm reminded heavily of the recent p-hacking controversy in science. Imagine a world where we "pre-register" a trial when an investigation is begun, and are then forced to go through with it (and pretend the government has the infinite money required to enable this.) In this world, that trial would likely declare the suspect innocent a lot of the time... and all the evidence gathered to do so would be "used up" by that trial. You couldn't turn around and use the same evidence to prove that "because he's guilty of X, it's more likely that..." anything else. Because, in the eyes of the law, he isn't guilty of X, and anything that proves that, doesn't. The evidence itself was nullified when the verdict was.
In the real world, just to save time and money, we don't bother to prosecute trials we know we'll lose. But shouldn't the effects still be the same as if we did? An optimization shouldn't change the semantics of the system.
The problem is that what he did /was proven amounted to a life sentence and the judge could have given him that.
Honestly I don't think judges should sentence based on what was not proven beyond a reasonable doubt. It is was so clear, why wasn't he charged for it?
Honestly I don't think judges should sentence based on what was not proven beyond a reasonable doubt. It is was so clear, why wasn't he charged for it?
Who's trying re-write history here?
https://web.archive.org/web/20160305020239/http://www.thegua...
You're omitting very important details.
https://web.archive.org/web/20160305020239/http://www.thegua...
You're omitting very important details.
Didn't he try to have five people killed only after they went after him?
I believe that he was sentenced for money-laundering, hacking, and conspiracy to traffic narcotics. You do yourself no service to memoize this as "running a free market."
> conspiracy to traffic narcotics
I don't recognize this as a legitimate offense to law in a free society. I see it as, yes, "running a free market". That the state chooses to serve its interests by finding differently is unsurprising. But why do you?
I don't recognize this as a legitimate offense to law in a free society. I see it as, yes, "running a free market". That the state chooses to serve its interests by finding differently is unsurprising. But why do you?
Can I open e free market for contract killers?
Lowest bidder and fastest killer would win
Imagine what a mess the assassination insurance industry would be given how much of a mess that the health insurance industry is.
US Liberals: We need single-payer hitmen!
US Conservatives: Do you really want European-style socialism assassins?
US Conservatives: Do you really want European-style socialism assassins?
We (most of us) take it for granted but I'm pretty sure quite a few people are scared to death, dead man walking due to drug deals gone bad, extortion, blood feuds, kidnappings and so on. I'm sure super rich people have these conversations, just we discuss finding a cheap apartment.
My view is that if someone wants you dead, it's just a matter or time, or price. Maybe the Pope and POTUS are out of reach, the rest if you can't finish them with a handgun you can with an Ar-15.
My view is that if someone wants you dead, it's just a matter or time, or price. Maybe the Pope and POTUS are out of reach, the rest if you can't finish them with a handgun you can with an Ar-15.
"Maybe the Pope and POTUS are out of reach"
John F. Kennedy would like a word.
John F. Kennedy would like a word.
As soon ans I'm done talking to Abraham Lincoln :)
Now they shut entire cities
Now they shut entire cities
There is an entire industry for kidnapping, ransom, and extortion insurance.
Honestly, most kidnappings are parent without custody snagging their own kid. Rich kid kidnappings are a thing though.
Honestly, most kidnappings are parent without custody snagging their own kid. Rich kid kidnappings are a thing though.
Word on the street is, there is a darknet site where you can bet money on what day someone's going to be assassinated. That way, a "volunteer" can bet on the correct day.
I really hope they called it 'Deadpool'.
Or you can short the stock related to the person about to have a bad day, but that leaves a trail unless done carefully. Even if done carefully, too many people know the secret. At least two, or one more
technically it was for hiring someone to murder a guy. Or are you implying that was entrapment?
edit: i stand corrected
edit: i stand corrected
On August 21, 2014, Ulbricht was charged with money laundering, computer hacking, and conspiracy to traffic narcotics. None of the procuring murder allegations were part of Ulbricht's indictment in New York, although the evidence was factored into Ulbricht's sentence.
https://en.wikipedia.org/wiki/Ross_Ulbricht#Silk_Road,_arres...
https://en.wikipedia.org/wiki/Ross_Ulbricht#Silk_Road,_arres...
No, it wasn't for that. He was never charged with that and it is likely that those messages were sent by FBI agents trying to frame him.
Or are you just trying to manipulate public opinion?
Or are you just trying to manipulate public opinion?
> He was never charged with that
It was explicitly a sentencing factor; the way federal criminal sentencing guidelines work, the charge sets the outer bound of punishment and influences the sentencing calculation, but other sentencing factors play a significant, often overwhelming, role in determining the actual sentence, so it's not at all inaccurate to say that a person is serving time for all of the things weighed in sentencing.
It was explicitly a sentencing factor; the way federal criminal sentencing guidelines work, the charge sets the outer bound of punishment and influences the sentencing calculation, but other sentencing factors play a significant, often overwhelming, role in determining the actual sentence, so it's not at all inaccurate to say that a person is serving time for all of the things weighed in sentencing.
I just tried to look this up and couldn't find a source that the alleged murder-for-hire was included in his sentencing. All I found was:
> On the last day of trial, Serrin Turner, the NY lead prosecutor, addressed the jury and stated that none of the six contracted murders-for-hire allegations occurred.
from https://en.wikipedia.org/wiki/Ross_Ulbricht
It seems like it would be unlikely to take the murder-for-hire allegations into account in the sentencing after the prosecution have already stated that they were false allegations.
> On the last day of trial, Serrin Turner, the NY lead prosecutor, addressed the jury and stated that none of the six contracted murders-for-hire allegations occurred.
from https://en.wikipedia.org/wiki/Ross_Ulbricht
It seems like it would be unlikely to take the murder-for-hire allegations into account in the sentencing after the prosecution have already stated that they were false allegations.
Wikipedia is a tertiary source; fact claims in it come in two forms:
(1) Ones that cite no other source and should be viewed as extremely unreliable, and
(2) Ones that cite some other source, which should be reviewed and evaluated to assess the claim.
The one you point to is in the latter category, and cites as it's support a Guardian article which describes be prosecutor not as claiming that the six attempts to procure murders for hire had not occurred, but that he did not believe any of the murders Ulbricht had attempted to procure had occurred.
(1) Ones that cite no other source and should be viewed as extremely unreliable, and
(2) Ones that cite some other source, which should be reviewed and evaluated to assess the claim.
The one you point to is in the latter category, and cites as it's support a Guardian article which describes be prosecutor not as claiming that the six attempts to procure murders for hire had not occurred, but that he did not believe any of the murders Ulbricht had attempted to procure had occurred.
> likely that those messages were sent by FBI agents trying to frame him
Do you have anything from a legitimate source on this? I'd like to read more. My understanding of this entire thing came largely from brief news reports on it.
Do you have anything from a legitimate source on this? I'd like to read more. My understanding of this entire thing came largely from brief news reports on it.
You can read about the corrupt FBI agents here: https://www.wired.com/2016/01/ross-ulbrichts-defense-focuses...
I recall reading that the FBI had admin access to the site at the time the incriminating messages were sent, which I'm still looking for a source for, so it's quite likely that the messages were sent by the corrupt agents. Particularly when you consider the libertarian ideology around Ross and the site in general. Soliciting a murder would be completely out of character.
I recall reading that the FBI had admin access to the site at the time the incriminating messages were sent, which I'm still looking for a source for, so it's quite likely that the messages were sent by the corrupt agents. Particularly when you consider the libertarian ideology around Ross and the site in general. Soliciting a murder would be completely out of character.
I don't think it's out of the question that he got in over his head and said "ok" to a guy offering to make the problem go away.
For what it's worth, the FBI agents on that case ended up going to jail as well. The case was a huge mess.
For what it's worth, the FBI agents on that case ended up going to jail as well. The case was a huge mess.
No genuinely thought that's what they got him for. Should read up on it I guess.
I love Tor and use it often (also collaborated with code and nodes). But the problem lately is that all sites are banning it. Even HN, gives you trouble when connecting through Tor.
Tor is often a vector for abuse, so tor node IPs get banned. I'd like to see a way to have nodes block certain origins without being able to identify them.
That being said, I really like the new onion-v3 protocol that Tor added as it moves from the outdated 1024-bit rsa + sha1 addresses with 1024-bit DH and AES-128 to the state-of-the-art ed25519 for addresses and handshake as well as chacha20-poly1305 for encryption.