50 comments

[ 4.5 ms ] story [ 83.6 ms ] thread
This is a tricky subject because all of us (excluding the pedophiles) can agree that sexual abuse of children in all forms is a bad thing.

That said, the erosion of privacy in the name of "think of the children!" is equally a bad thing.

In this case it appears to be a legitimate case of a sexual predator sharing his digital trophies and rightfully being caught.

But "protecting the children" doesn't have to mean giving up our expectations of privacy in the normal course of affairs.

Sorry for being pedantic, but pedophilia is psychiatric disorder. I'd imagine all pedofiles WOULDN'T think child abuse is good.

That said, the word you were looking for, is child molester. People need to realize that not all pedofiles act on their attractions, nor that all of them would want to. It's important to note that the feelings of attraction, do not automatically mean they want to rape all children...

You just labeled yourself a child rape apologist by social justice warrior feminist standards.
Well to me, I'm simply highlighting the fact that not ALL pedofiles are rapists.

It doesn't surprise me that some/most people believe this, but I believe that defending simple truths to be important.

Again, pedantic. But necessary. Especially if people who go by social justice warriors, believe in the persecution of pedofiles who've never commited or will commit a crime.

>This is a tricky subject because all of us (excluding the pedophiles) can agree that sexual abuse of children in all forms is a bad thing.

More like the society doesn't allow discussion on these topics. Any opinion which is not perfectly in line with the "official truth" is not acceptable, so there's no point to talk about it.

^ Pretty much this.

However, as much as it would be great to have the freedom to talk about specific topics. I am moreso concerned on figuring out why Google gives a shit, and what else they are hand-delivering to the feds.

Looks like it's reformat, and use a different search engine time.

So, uhm, you think there are large numbers of people who think that sexual abuse of children in all forms is a good thing? And that they don’t dare speak?

I sure hope not! Or what was your point again.

The point was that no points can be made.
>is equally a bad thing.

You really think that rape of children is equivalent to telling law enforcement about images of that child abuse?

There's a mild concern here, but it does rely on two things - having a known image and having it uploaded to a web service. Judging by the information in the article an original image wouldn't be matched, and it is in the ToS (good luck reading all of that).

The organised groups that make money off this are much wiser and harder to catch.

More interesting is the quote "Microsoft also developed Photo DNA to match pictures using just a pixel". I'm not sure it works quite how they suggest.

The Picasa policy spells out that they don't allow those kinds of things. I would assume one way they would check would be some kind of algorithm. I see "no evil" here on Google's part.
Yeah. But a companion question should also be - if they use a potent technology for something i approve of (and we all approve of that uses) can it be (ab)used for something we dislike.
Sure, that's only reasonable. But you're at that point only one step away from a slippery slope argument. (How meta..)

Problem is it's not always a valid question given the context. Automagically scanning images with an algorithm to find known sexual abuse pictures is in a completely different world than scanning images with an algorithm to find copyrighted content, for instance.

I think this is a crucial point. To make it more concrete:

In the U.S., having a video of Osama bin Laden congratulating the 9/11 terrorists might be grounds for suspicion.

In China, having a picture of a guy standing in front of a tank in Tiananmen Square is highly frowned upon.

The least you should expect when you upload something onto a cloud service is that your data will be subject to all kinds of algorithmic analysis.

And it's just not tenable for Google to not do this kind of matching. Imagine the flak they would get if they refused to match against a database of child porn citing "privacy concerns." Such a refusal would be neither pragmatic nor ethical.

From how I read it: Google match all images to a database of file hashes.

I wonder whether they do it for redundancy purposes like Dropbox, or whether it's entirely to crack down on child pornography.

If they only did it to remove redundant pictures, they would have no way of knowing that a given image was porn; all they'd know was that it was identical to some other image on their servers.
Considering this person uploaded explicit pictures of children to a cloud server, is anyone actually surprised that the FBI showed up at his door soon after?

In addition to what federal and international law says on the matter, basically every web service on the face of the earth explicitly forbids these kinds of images in their terms of service. Once he hit the upload button, this guy's 'privacy concerns' were kaput.

Slightly offtopic:

"Microsoft also developed Photo DNA to match pictures using just a pixel."

What. So if I give Microsoft a pixel with value #F00 it can determine if that's from a specific photo?

Great reporting with great insights and sources.

Ya that's ridiculous but it would be hypothetically possible if you had a sufficiently high number of colors. Then the chance of two images having a pixel with the same R, G, and B values all exactly the same would be exponentially small.

You could possibly do this by averaging all the pixels together for example.

I don't know how I feel about this. Merely looking at CP doesn't actually hurt anyone, and pedophiles are not necessarily attracted to what they are by choice. But there is an argument that it's existence encourages or can lead to child abuse in the real world, especially the production of more CP.

This seems to be demonstrated by this case. They caught the guy through CP but it turns out he actually abused a kid. Maybe he wouldn't have been caught otherwise.

On the other hand is punishing them and throwing them in jail for years actually helpful? As opposed to a lesser sentence or rehabilitation of some kind maybe?

Another problem is that it's incredibly easy to frame people for. Just put some cp on someone's computer or Google account and you can have them sent off to jail with little effort and no way for them to prove they are innocent.

Child PORN should not exist for people to look at.

I take great issue with your first sentence.

Who the hell do you think was harmed in the creation of said CP? If you are in any way sympathetic to those that would "harmlessly" watch it, then you are tucking delluded as to the reality of such a situation.

Seriously, take a moment to think about what you said and try to not pretend to be so "academic" about your position.

I'll assume you have no children.

Do you mean that looking at images harms people, or that people choose to be pedophiles?
I take issue with it's creation, but once it exists, looking at it is a victimless crime whether or not you find the act objectionable.

Which is another reason to be skeptical of the laws as they main reason they exist is because people find it objectionable, rather than rational arguments about what actually prevents child abuse.

>looking at it is a victimless crime

It creates a market. Is there no empathy for the subject of the photo? Would you like your 6-year old son or daughter to be looked at with a sexual gaze?

It would only create a market if the recipient paid for the porn or exchanged some other porn for it. I agree that participating in a market for child porn should be a criminal act. However, I don't see how merely possessing an image that you happened to find on the web can create a market (or cause harm to the person depicted in the image).
Victimless crime?

How about you ask those who were abused whether they feel like they are not victimised by people spreading and owning images of their abuse.

The victims are quite obviously real.

For the most part the victims are anonymous and will never know whether the photo is spreading unless they themselves go to the darker places of the internet.

I see your point, but looking at a photo of someone without their permission is vastly different than raping them.

So you think it should only be a crime if the victims know about it? That seems like a weird and unnecessary wrinkle. And I don’t think most victims have to go to “darker places of the internet” to know, given that they were present when the abuse took place and was filmed/photographed.

I would also like to know when I said that raping a child and looking at images of said rape are equivalent crimes.

Also, it’s looking at an image of someone being abused, not just an image of someone.

Are all CP images images of abuse? For instance, if two teenagers film themselves having sex, are they both abusing each other?
Photos of abuse are by definition proof that abuse happened and there's a very strong argument that demand creates supply.

It gets trickier if the images merely misappropriates existing images but there's still something fairly creepy about that.

However - there's also a move to outlaw synthetic images which is more troubling. I recall a case where someone was prosecuted for possessing pornographic representations of "The Simpsons" or something similar

I feel it's certainly problematic if all three categories are treated identically as the amount of direct and indirect harm differs significantly in each case and the tradeoffs we make regarding privacy vs. protecting individuals from hard likewise would be different in each case.

And now go to a library and burn a copy of Lolita. And Song of Fire and Ice while we are at - Deneris was only 12 in the books.

The current legal landscape is that something could be CP even if no real human beings were used in its production. Or were over 18 but looked too young, or they were a 17 year old kid that send someone a picture of his dick/pussy as joke via snapchat.

Saying that something that is so vaguely defined shouldn't exist gives a free victory for censors.

Disclaimer - I suppose that you understand under CP what us normal people understand and what an overzealous DA, but still.

>burn a copy of Lolita.

No children were harmed or molested in the making of literature. That's a different topic to actual, real-life photographic CP.

> No children were harmed or molested in the making of literature. That's a different topic to actual, real-life photographic CP.

That isn't what the US statutes use to differentiate between legal and illegal, however; rather the Miller test is what applies.

Should videos of terrorist attacks also not exist? People were harmed in the production of said videos.
We don't want to see people who do things we don't like as people. We'd rather put a label on them and ship them away.

How does it make any sense for a woman who stops to pee near some bushes at night get on the list just because where she did it happened to be next to a closed park where there were no children at that time but there may have been kids during the day?

Nope, just put her on the sex offender's list and make sure she has to stay xx miles away from children.

>I don't know how I feel about this. Merely looking at CP doesn't actually hurt anyone, and pedophiles are not necessarily attracted to what they are by choice. But there is an argument that it's existence encourages or can lead to child abuse in the real world, especially the production of more CP.

That argument applies to virtual cp too.

Law Enforcement have been using the hashes of known child pornography images to find pedophiles for probably a decade. This technique is no different than what anti-virus vendors do to detect malicious software. I'm starting to hate Google in lieu of their new campaign on killing privacy and anonymity but I don't see the privacy concerns here. Server side hashing of images and comparing them to a database isn't exactly gives metadata to the NSA. I think it's good that Google is doing this. Odds are more image hosting providers should be doing. It's low hanging fruit but at least it could protect a child.
Is a matching hash enough to get a warrant or is a warrant even necessary? Or is it probably cause to raid the user's house at that point? I'm curious to know the process once a matching image fingerprint is found.

Or will we soon be hearing of a police raid on a person's house because of an errant MD5 collision?

According to US law, any online service provider that receives actual knowledge that child porn exists on their services is required by law to report it to the National Center for Missing and Exploited Children... or face a fine of $150k for a first offense, or $300k for subsequent offenses:

http://www.law.cornell.edu/uscode/text/18/2258A

To my knowledge, Google wasn't required to check its hosted files for specific image hashes: "Nothing in this section shall be construed to require an electronic communication service provider or a remote computing service provider to— (1) monitor any user, subscriber, or customer of that provider; (2) monitor the content of any communication of any person described in paragraph (1); or (3) affirmatively seek facts or circumstances described in sections (a) and (b)."

But once they did, they were required to report the results. Any privacy concerns are waived as an "[exception] for disclosure of communications" as part of 18 USC § 2702: "Voluntary disclosure of customer communications or records". http://www.law.cornell.edu/uscode/text/18/2702

It's not surprising that Google is affirmatively scanning its servers for child porn, as they have a long history of providing assistance to NCMEC:

http://googleblog.blogspot.com/2008/04/building-software-too...

http://googlepublicpolicy.blogspot.com/2012/01/crowdsourcing...

"$1 Million from Google to Fight Child Sexual Exploitation Received by the National Center for Missing & Exploited Children" http://www.missingkids.com/news/page/4898

The question is what else are they scanning for, to give to the FBI/police.

I remember Facebook announced a while ago that it would scan its Facebook chat for "criminal activity", to give the info to the police.

http://news.cnet.com/8301-1023_3-57471570-93/facebook-scans-...

Where is the line drawn? And will we know where it's drawn? Stuff like this makes me trust "cloud computing" less and less. I hope Google and others realize this, before it's too late (for them) to change course.

I can take a guess where this is going, though, if NSA's actions are any indication - there won't be a line anymore, if they (government/corporations) are left alone to continue with stuff like this. Every single crime, no matter how small and insignificant, will be reported to the police, thanks to the mass surveillance capabilities of the alliance between NSA/government and corporations. After all, when "safety" is the ultimate goal, above all else, that outcome is all but inevitable.

It's already started:

http://falkvinge.net/2013/11/24/nsa-mass-surveillance-has-al...

This isn't where the line was crossed. Far from it. Even though eavesdropping on the everyday citizen is a pretty damning thing to do, in their eyes it isn't illegal so they can't understand why everyone is complaining.

The line was crossed a long time ago when the Law was treated as an opinion. This cultivated a "let's see what we can get away with" attitude which is now starting to unfold before the world stage.

For more info, see:

  https://en.wikipedia.org/wiki/John_Yoo
Seems incredibly easy to frame someone for child porn.
“The state must declare the child to be the most precious treasure of the people. As long as the government is perceived as working for the benefit of the children, the people will happily endure almost any curtailment of liberty and almost any deprivation. ” -Mein Kampf, Adolf Hitler, Publ. Houghton Miflin, 1943, Page 403
This doesn't make me significantly worry about my privacy using Google services. Google knows that it needs to maintain trust with its users and that's why it's only tackling a crime which has little controversy with respect to its heinousness and is almost trivial to detect, from a systems perspective. I'd far rather Google show good faith and help fight crime on their terms, rather than the government come in and start imposing safety regulations and make it obligatory.
Wow, there's some fucking dire opinions being expressed in this thread. How fucking removed from reality are some of you.