Hacker News new | past | comments | ask | show | jobs | submit login
80 Percent of Tor Traffic Is Not Child Porn (wired.com)
224 points by Libertatea on Jan 28, 2015 | hide | past | favorite | 58 comments



The claim is nothing more than propaganda. The DOJ knows the truth, and the people who are part of the technical community know the truth.

Saying this is just using false information to sway the public, and looking to badge things that are just tools as inherently evil tools, or tools only used for evil.

Disclaiming the lie is a good step - but I'd like to see the rebuttal coming from a more mainstream media source. Wired writing this article is preaching to the choir.


It's the "save our children" argument -- the very same argument the UK's Cameron et al are using to propel new legislation like banning all non back-doorable encryption, and so-called "porn filters" a la blanket censorship of whatever the gov't deems censor-worthy (sometimes even political content).

They spread fud and fearmonger the general population into believing something must be done to prevent all of this "law enforcement circumventing" encryption and anonymity (aka. privacy)...

You know... because only criminals use encryption and anonymity tools. Only criminals have anything to hide; that they don't want other individuals and criminals (and government) to obtain.

Forget all the good and valid reasons for encryption and privacy! Like protecting your medical records... protecting your financial records... your private correspondence... or the very ability to enter your credit card information on a website and trust it can't be MITM'ed. Only criminals want any of that.

It's rubbish to say the least.


In essence it's a false flag operation. They are having problems getting support for corrupting the Tor network so they are making things up.

I wish someone in Congress would ask and investigate the DOJ and FBI on how much of the kiddie porn is actually generated and associated with and originating from DOJ and FBI operatives who are trying to attract those who would want that kind of thing. I know it's not 0.


I have a tough time imagining that they would generate (taking this word to mean "create") child porn. I could easily see them taking what they've found and publishing it however.


I gather that various conservative cultures see much of Western popular culture as child porn. Even to arguably dissolute Americans, aspects of popular Japanese culture look a lot like child porn.

Who gets to set standards on the Internet? If the least-permissive standard won for each category, the Internet would be boring.


I have absolutely no difficulty imagining the ruling elite using their heinous secret society to produce material that can be used to repress us all. One only need look at how well the Franklin Affair was managed to see that the elite do indeed have their inner society, and that they can get away with the most heinous crimes, given enough secrecy and deceit; something our ruling masters are getting better at as the years go by.

Most damaging, I think, is the ability of governments to keep secrets - this is the most destructive force in governance, I believe, and we the people have to realize that the right for governments in power to keep secrets is a form of violence upon the body politic. But unless we yank the chain by revoking that power on occasion, I think it gets abused all too well.

In fact, I think the only valid response to the governments new calls to remove public encryption is to demand that the right for governments to keep secrets be revoked, by the people. Let us answer their call for intrusion into our private lives with the response: no more secrets in government then, either!


>Most damaging, I think, is the ability of governments to keep secrets

Hear, hear! But propose that the government be open and transparent and people complain about all of our military secrets being public. Fucking good, that's the point.


> I could easily see them taking what they've found and publishing it however.

Which in itself is illegal.


Of course, but is also different than what was alleged.


Having worked with attorneys (including some at the DOJ), I would advise you not to overestimate their understanding of technical concepts.


The source of the confusion here is obvious and pounding on it won't help. From the DoJ's perspective Tor is hidden services. Criminal abuse enabled by proxying to the open web is not high up on their radar. Regular web sites can easily block Tor. But hidden services are unique. When Tor is mentioned in the media, it's almost always because of hidden services. When it's involved in large scale FBI investigations, it's almost always because of hidden services.

So whilst Andy Greenberg's article is correct, it's correct in a nitpicky technical sense that isn't going to have any political impact as an argument. Tor has chosen to support hidden services very strongly, so the fact that it's a relatively small amount of their total traffic doesn't matter much.

The Tor project needs to take this kind of thing seriously. Alarm bells should be ringing. They are getting (apparently credible?) tipoffs about plans to seize directory authorities. Relays actually are getting seized. DoJ officials are publicly saying they perceive Tor as being dominated by crime. Meanwhile they are doubling down and making blog posts that describe police operations as "attacks". This has the feeling of a slow motion train wreck in the making.


What exactly do you expect Tor to do about it? Say what you mean.

Hidden services have a wide range of legitimate, valuable uses, including: https://firstlook.org/theintercept/2015/01/28/how-to-leak-to...


> that describe police operations as "attacks"

Is there a more reasonable way to describe "police operations" whose goal is to disable portions of or function of a network?


Good point. It is the standard term in security and cryptography.


It doesn't matter. You can't use in-group specialized vocabulary when talking to out-group audiences. You must either use common vernacular or somehow give new, commonly accepted definitions to the words you want to use.


I don't think "attack" is specialized vocabulary meant for an in-crowd.

For reasonable, common definitions of "attack," seeking to disable something or cause it to malfunction is an attack against that thing.


Did anyone else initially read this title as implying that 20% of Tor traffic IS child porn?

Edit in case it's changed: The title is, "80 Percent of Tor Traffic Is Not Child Porn."


Unsurprisingly, the original title, "No, Department of Justice, 80 Percent of Tor Traffic Is Not Child Porn," is quite a bit clearer.


What about the actual url...

`/department-justice-80-percent-tor-traffic-child-porn`


The internet community as a whole is going to have an uphill battle against telecoms, spy organizations, and law enforcement agencies ...probably forever.

I hope that at some point John Q Public realizes that digital privacy and free flow of information are more akin to advancing society scientifically and spiritually much more so than terrorism, child porn, human/drug/weapon trafficking, et al.

After the Snowden revelations it was quite apparent that the average citizen would gladly exchange a lot of privacy in order to fight an enemy that doesn't exist.

I hope we can stop going to war with ghosts and focus on moving society forward. An open internet can be a great tool for this.


>>I hope that at some point John Q Public realizes that digital privacy and free flow of information are more akin to advancing society scientifically and spiritually much more so than terrorism, child porn, human/drug/weapon trafficking, et al.

The trouble is, there is no way to accurately measure this factor, nor is there any recognition of it - because using the Internet for free and open, honest communication, is the majority case. Unless we can come up with some way of acknowledging the good, and really getting the good counted in a way that is statistically significant in the discussion, this piece of the pie is going to be impinged upon by the loud, attention-getting, criminal activities.

But this circumstance is not unique to the Internet - it has been true of all media, since the days of Gutenberg. The good rarely gets attention; only the violence, the hatred, the terrorism gets real attention in our world. Nobody turns on the news to hear what good has happened in the world - for some reason, this is not interesting to the majority.

If there were only some way to change this situation, we would see some light at the end of the tunnel - I suppose acknowledgement of the positive in the world does happen. But we need to get off our junky fix for the disorder and chaos in this communication age, and figure out ways to measure the good that open, honest, and most importantly free communication brings to the world. We need a metric that makes sense in light of the death statistics, the crime rate, and so on ...


>The internet community as a whole is going to have an uphill battle against telecoms, spy organizations, and law enforcement agencies ...probably forever.

Wishful thinking. Way things are going we will lose that war soon.


The average citizen doesn't care as long as they get their bread and circuses.


Well the average citizen cares about certain things such as "jobs are good" or "terrorists are bad" where they don't necessarily care but they at least make associations with certain topics. I guess I hope that privacy/open internet will become one of those simple things people can understand.


I made clear at the time that the study claimed 80 percent of traffic to Tor hidden services related to child pornography, not 80 percent of all Tor traffic.

This is also a bogus number. An estimated 80% of Tor hidden service directory lookups are for those sites. Long-lived connections only make one lookup, and sites that people check multiple times in one session only get looked up periodically. The ones that are looked up the most often are the ones where people open Tor, visit the site, then close Tor. This does not correlate with volume of Tor traffic at all.


It probably correlates really well with random research projects that are tracking those hidden services...


It is only natural that people with things to hide will be overrepresented among early adopters of technologies for hiding things. The solution isn't to weaken hidden services, so that evil can be more readily exposed. Far more useful will be increasing the accessibility and usability of hidden services for protecting human rights. In this case, dilution is truly the solution.


And while we're discussing it, 80% of people carrying cash are not drug smugglers, and 80% of people using Waze to watch for traffic cops are not cop-killers. Are we seeing a trend here?


Wait, is that actually a criticism levelled against Waze? That can't possibly stick, right?

Almost everyone in the U.S. drives and everyone who drives hates getting traffic tickets. I thought allegations/criticisms like these only really worked when you could separate the users into some group identified as "other".


Yes that is a real criticism against Waze. The thought is that the "show where the cops are" feature can be used by those who wish to assassinate cops (by finding them when they are alone I guess).

The obvious conclusion being that the feature must be removed (obvious to a certain Orange country sheriff at least). http://www.npr.org/blogs/thetwo-way/2015/01/28/382013185/off...


There is a movement (it's currently trending on Facebook - so there are plenty articles to read there) that is trying to force Google to remove the offending data. The problem is, to anyone's knowledge, there hasn't been a single case of "officer assault" linked to Waze data.

It does sound silly, as cops are not hard to find. I seriously doubt a would be cop killer needs to consult an app when looking for a cop to kill.


> cops are not hard to find.

IIRC they even have a common phone number to summon them.


That's not saying that "80% of Waze users are cop killers" though. Just that some people could use it to locate cops in order to attack them.

I don't know if that makes really sense but it's much less outrageous than what the OP implied.


I think what the OP implied is that things will always be used by criminals for other purposes than intended.


I think it's more likely to be used on the commonplace to avoid areas with police -- either for the traffic ticket dodger, or for mugger's, home invader's, etc...


> feature can be used by those who wish to assassinate cops

Are you trying to get a government job as propagandist and fear-monger?


No he's describing the reasons that the police spokesmen are presenting as to why Waze should remove the police observation data.


http://hosted.ap.org/dynamic/stories/U/US_POLICE_TRACKING_AP...

From the article: Los Angeles Police Chief Charlie Beck complained in a letter to Google's chief executive on Dec. 30 that Waze could be "misused by those with criminal intent to endanger police officers and the community."


Perhaps then the police should stop tracking everyone's movements with Automated Number Plate Readers (ANPRs) because the second their huge database gets compromised it's going to be exceedingly easy for criminals to figure out when law abiding citizens are regularly not home (and not just during work hours!) so they can safely break into houses and rob them.

What's good for the goose is good for the gander.



Wouldn't you have to decrypt and inspect all Tor traffic to accurately determine what percentage of it is anything?

If the DoJ could do that, they would necessarily have the ability to track and capture the CP-traffickers without bothering to "inform" the news media that such people are crawling all over Tor, in your Internet pipes, coming to get your children.

CP is such a universally reviled thing that whenever someone brings it up in an unrelated argument, I automatically suspect that they are losing it, they know it, and they are falling back to rhetoric. It is the "won't someone please think of the children" appeal to emotion, turned up to 11.


How would anyone really know? Isn't the point of Tor to make that kind of study damn near impossible? For all we know 80% of Tor traffic is hot sexy manatee action.


Unencrypted data passing through an exit node can be read by the operator of the node. Deploy x amount of nodes, analyze traffic and perform some basic statistical analysis?

Granted your data will be biased towards unencrypted traffic, but you should get a small picture of whats going on. Plain jane http and ftp traffic should be sniffable.


There is no amount of statistical analysis that you could do to extrapolate the amount of traffic with anything even approaching accuracy. I also have approaching zero confidence that any government agency or even a contractor funded by the government would be able to approach adequate methods to allow even an attempt at statistical analysis.

The "small picture" you would get, is very well probably the 80% we are talking about, e.g., 80% of all traffic that is associated with kiddie porn and is also unencrypted because the perpetrators are unsophisticated is actual kiddie porn traffic, which says nothing whatsoever about the overall amount of traffic that is kiddie porn. On top of that, what are we talking about here? What is this definition of o kiddie porn? Does that number include subjects that "appear" to be underage? If I know anything, I know that the government will skew numbers to fool others and even themselves; because that's how you get money and how you build your little fiefdom.


That doesn't tell you anything about the hidden services. By design, the hidden services don't have exit nodes. The traffic is encrypted from client to server.


The statistic in question was in regards to requests to hidden services.


99% of the population does not check the numbers that governments and media put out.

Also, governments and media are notorious for giving out numbers that are nearly impossible to find out anyway.


At this point, it's not outside the realm of possibility that some or many of the hidden services are being run by three letter agencies themselves.


These agencies used fronts to donate money to cold war cultural institutions. I think your speculation is entirely possible, even though it sounds tinfoil-hattish now.

The Cold War example that comes to mind is Partisan Review, a broadly left-oriented journal that published Susan Sontag's "Notes on Camp" and Clement Greenberg's highly influential art essays (http://en.wikipedia.org/wiki/Partisan_Review#Funding_by_the_...), which (it was later found) accepted multiple donations from CIA fronts.

As time went on, some viewed these donations as a tool these agencies found to bolster the intellectual stature of the "New York School" of heroic AbEx painters versus their Soviet counterparts (http://en.wikipedia.org/wiki/Abstract_expressionism#Abstract...).


Has anyone actually seen that "study" from Portsmouth University?

I tried to google it but all I get is useless soundbites and fear-mongering, all over the mainstream media, with everyone linking back to Wired.com.

No one links to the actual study or to any evidence to support the claim.

Shame on you Wired.


The statistic comes from a talk at this year's 31C3 In Germany and was delivered by a professor from the University of Portsmouth -- Dr. Gareth Owen.

His personal site: https://ghowen.me/other.html

And a direct link to the video: https://www.youtube.com/watch?v=oZdeRmlj8Gw

I'd imagine he'll publish a full write-up soon but the talk was only a few weeks ago so il not surprised it's not out yet.


How do we know how big a percentage of Tor traffic is child porn? Who logs that traffic?

Also, this is definitely one of the reason I can't see myself using Tor. If I were to be a proxy for something like this :(


While the DOJ should, but probably won't, correct it self, and should have gotten it right the first time, Andy Greenburg deserves a significant amount of fault for this misinformation. Both articles use somewhat ambiguous language, and have titles that are unclear.

The most important distinction is between "Dark Web" and "Tor". A reader not intimately familiar with Tor could easily assume that those are used interchangeably.


How would it be relevent even if it was? You either accept the fact people are free or you don't. Whatever is your judgement on a specific topic should not matter when talking about tor. Be it child porn, human traficking, banned journalism.

Why would US's judgement against child porn be better than <insert a country>'s judgement on journalism.


I've heard it said that the way to win a debate is to show the opposing position ultimately leads to genocide, nuclear war, or child pornography. It's depressing to witness the government's fear momgering reach its nadir.


Beware. This is how they shut down Usenet.


Who writes these headlines? 80% of the TOR traffic is not-child porn, so only 75% is? Shouldn't the headline mention how much of it is child porn(its close to 1%, from the article).


Usually it's an editor, not the author, who picks headlines. Some sites even A/B test their headlines to see which gets more clicks.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: