Sure, but then we're not talking about the vast majority of digital transgressions are we? That is, the simple (re)distribution of child pornography — ghastly enough as that is — not the few who actively seek out and order new abusive material from those who actually produce the stuff.

This debate is about proportionality. How many children can live a marginally better live by giving up strong encryption and privacy worldwide? Because with the exception of the few cases in countries with a strong and well-functioning legal system, the abuse doesn't stop when the on-line orders stop coming in. How many activists and oppressed minorities will lose out (livelihoods and/or lives) by doing this?

And even then, you're not addressing the real problem with this anti-encryption stance: how are you going to prevent the bad guys from using encryption? Encryption is here, right now, and it exists as a mathematical fact. Introduce a government (incidentally, which?) backdoor in WhatsApp or whichever major platform, and you catch some low-hanging fruit, until they wise up and use the vetted tools without a backdoor that already exist to hide their activities. This significantly impacts the proportionality balance.

If you want to prevent child abuse, investing in education, welfare, economic equality, respect for basic human rights, and boots-on-the-ground police capacity will net you a vastly better return on investment.

I'm not at all trying to pull this into the 'think of the children' domain, I'm just pointing out that what you think is going on is probably not nearly as bad as what is really going on. It goes a lot further than just child pornography, think snuff movies involving children made to order and other 'creative' idiocy.

I'm all for people having access to encryption, I'm also for stamping out these criminals. The two are not mutually exclusive.

I have no doubts that whatever depravity a child can be submitted to that I can think up has or will have happened, and worse. There is certainly no god that prevents it, there is only us humans, and there are a handful of very sick individuals among the seven billion — that's just a statistical surety.

> The two are not mutually exclusive.

There we agree. But politicians worldwide do keep turning this into a 'think-of-the-children' fight. Only this month the Dutch minister of justice and security came up with exactly that argument for forcing Facebook to include a backdoor in WhatsApp — a plea indubitably neatly prompted by his colleagues in the US (cf. attorney general Barr's position on this topic).

> and there are a handful of very sick individuals among the seven billion

This is where you're wrong. Child porn is a substantial market, estimated to be in the billions of $ per year. It's a disproportionate amount compared to 'regular' porn because of the criminal element and the risks involved, just like illegal drugs are far more expensive than legal ones (in most parts of the world, anyway).

Politicians abusing this to push their own agenda's are effectively also abusing these children... it's a fucked up world, that's for sure.

Billions of dollars per year? On a world basis, that's quite insubstantial.

And child abuse has always been around, unfortunately. Much of it probably goes unrecorded. Indeed, we only became aware of how bad the problem really was in the 1980s, due to the very fact that media of it was now being created and exchanged.

I aspire to one day reach your level of cynicism but for now I will have to disagree with you that that is insubstantial, even on a world basis. Especially so because it is concentrated in only a very few countries where children have a very high chance of being exploited in this way.

Have a read.

https://www.rappler.com/nation/191219-philippines-top-global...

With respect, which part of this article do you find interesting? In fact, it totally confirms my expectation going in, that the whole thing is empty PR phrases and bullshit numbers. I mean, take the very first line:

"The Philippines has become a top global source of child pornography with around 80% of Filipino kids at risk of online sexual abuse or bullying, a global Unicef report said Tuesday, December 12."

That's the definition of bullshit. What is the actual claim here? 80% at risk (so maybe nothing happened). Of what? Well, possibly "bullying". How is that related to child pornography?

Ignoring the fact that the so-called "at risk" numbers are a favourite way to inflate the danger, because you have very wide latitude to make that number up, this claim is particularly incoherent.

I can't even, in this article, find a solid claim that child abuse is a particularly wide-spread problem in the Philippines; never mind any evidence for such a claim.

The article quotes a Unicef report that I can't find a good source for, and the Phillipines being problematic in this respect is something that is not exactly a secret, it is where a lot of 'production' happens. I'm sure if you dig around a bit you'll find lots of corroboration for that datum. Anyway, I'm on the clock and only look at HN during breaks right now, if not for that I'd be more than happy to find better references.

> The article quotes a Unicef report that I can't find a good source for,

And what does this Unicef report say?

> and the Phillipines being problematic in this respect is something that is not exactly a secret

Terms like "problematic" are excellent for persuasion, as they exclude any quantitative component as they invoke fear.

> I'm sure if you dig around a bit you'll find lots of corroboration for that datum.

Like the fear mongering article you linked?

Look, I'm not advocating for the abuse of children, but when I'm being encouraged to worry about something, I prefer the warnings to have some basis in fact.

Minimizing the incidence of digitally-enabled child abuse is not a winning strategy for protecting our right to use strong encryption.

For one thing, it puts you in the position of defining what incidence of child abuse is low enough to be somehow acceptable. Is the Internet helping one person to abuse a child? Any normal person would say that even one is unacceptable.

For another thing, it can easily make you look like an unserious fool because the people you're arguing with--law enforcement--are going to know way more than you do about the actual incidence of these particular crimes. They are privy to every ongoing investigation, and are generally not legally permitted to tell you about them.

We don't need to doubt the good faith of law enforcement--at least on this issue of child abuse--to advocate against encryption backdoors.

Encryption backdoors are a bad idea because they can be later abused or hacked, causing untold harm.

And we get to use strong encryption for the same reason we get to deny the government placing cameras in our houses. We get strong encryption for the same reason we get to be considered innocent until proven guilty. Efficacy or efficiency for law enforcement is not enough, and should not be enough, to supersede our rights to privacy, expression, assembly, etc.

> Is the Internet helping one person to abuse a child? Any normal person would say that even one is unacceptable.

What does "unacceptable" even mean? How much do these normal people think society be willing to invest to prevent that one instance of abuse?

For an extreme example: it's probably below the $10 trillion cost of completely shutting down the global internet.

> I prefer the warnings to have some basis in fact.

What would you consider a fact? A half dozen links to such content?

Really, I'm not sure what you are getting at here, there is plenty of evidence for this stuff being real and being sourced from a limited number of countries if you want to argue the opposite then I'm perfectly ok with that. But that Unicef report is there for you to read and if that does not convince you likely nothing will.

Just like NL has a - justified - bad reputation for hosting a lot of this content so the Philippines have - unfortunately - a bad reputation for being on the producing end. For a while it was Thailand, then the Thai really cracked down on the sex tourists and the child porn production and so the producers shifted to another location.

> What would you consider a fact? A half dozen links to such content?

Well, one link would be a start. And to consider something a fact, I'd generally want to see some evidence. When I'm told something is a fact but there is no evidence to support the assertion, or if the evidence turns out to be an estimate with questionable methodology methodology (say, 1 In 5 College Women Are Sexually Assaulted), then I tend to lower my opinion on the honesty of the individual or person involved.

> But that Unicef report is there for you to read and if that does not convince you likely nothing will.

It is?

> For a while it was Thailand, then the Thai really cracked down on the sex tourists and the child porn production and so the producers shifted to another location.

I don't follow Thailand terribly closely, but I'd be interested to know details about this, particularly how they know child porn producers were there but left.

> Especially so because it is concentrated in only a very few countries

The meaningful question to ask is what's going on in those countries, that makes abuse so much more likely to occur there? It can't just be weak government, the Philippines aren't generally known for having weak or ineffective government.

> estimated to be in the billions of $ per year.

That seems ridiculously high unless it's millionares being charged $500k a pop. Got a source on that?

I find it strange to see the same comment being replied to by that the amount is insubstantial and that it is ridiculously high. Maybe the two of you can come to some kind of an arrangement?

Anti-encryption stances will also concentrate power, and thereby make for less stable societies. That in turn will harm the rate of economic development, which is what brings children out of poverty where they are most vulnerable to child trafficking rings.

It's easy. Someone uses encryption? Jail him regardless of the content. You can't hide the existence of the message.

> Someone uses encryption? Jail him regardless of the content.

Right, so by using Hacker News you're falling afoul of your own rules.

> You can't hide the existence of the message.

You totally can: properly encrypted data is indistinguishable from random noise.

These are not my rules. The parent poster said that the government CANNOT ban encryption. It can. No message can be totally hiden from a motivated enough opponent.

The random noise IS the point. you cannot hide its very existence, and a government agent might not want to believe you if you said that it's just random data.

Great. Might as well just jack up the consequences to the point that perpetrators realize they have everything to lose and guarantee there is no low-hanging fruit, and that whatever you do end up having to infiltrate will have a "like-your-life-depends-on-it" security posture.

This is how cures become worse than the disease. Life isn't black and white, nor do things tend to be doable without gratuitous unintended consequences.

Encryption is not, nor has it ever been the be all end all to child predation. Effective investigation techniques have been the issue, and awareness have been the problem, because of the ephemerality of any distribution system once its existence is outed to hostiles in the non-pedophilia community at large.

If you want to take these people down, you have to map the logistics. If you want to map the logistics, you have to consume content; once you have the content, you need to analyze the crap out of it for every possible clue you can about how and where it was made in order to try to make some reasonable guesses about how the production industry works, then you have to double down and go deep to figure out what these production rings look like, whether they know each other, and how do they work. I'm talking picking apart the audio for environmental noise that might clue you in, or developing profiles of certain quirks of equipment. Analyzing the medium of exchange to see if you can exploit the financial traceability aspect, etc... All of which can also be sanitized once the opponent knows you look for it by the way. Proper OPSEC on their part will leave as few breadcrumbs as possible for you to find, and the closer you get to the heart of the op, the more heinous the infiltration will likely get.

All of that has zilch to do with e2e encryption. That's all meatspace work. Ugly work, but work that'll need to happen nevertheless.

What about /dev/random? You're gonna ban PRNG's while we're at it?

And HTTPS?

Of course you can hide the existence of the message. The art and science of doing so is called steganography. Some forms of steganography are intended to be completely undetectable, while others merely offer plausible deniability.

https://en.wikipedia.org/wiki/Steganography

from your very own link:

> Generally though, there are many techniques known to be able to hide messages in data using steganographic techniques. None are, by definition, obvious when users employ standard applications, but some can be detected by specialist tools. Others, however, are resistant to detection - or rather it is not possible to reliably distinguish data containing a hidden message from data containing just noise -

You cant hide the existence of the data. You can pretend that it's just random bits. But they are still here, and an opponent can see them.

> No, but you can order physical abuse over the internet

Is this actually taking place at a non-negligible scale?

Large enough that it bothers me. The Philippines is particularly problematic, the unofficial figures are that you're looking at roundabout 700K perps world-wide and an (obviously) smaller number of producers, how accurate those are I have no idea. Fortunately my exposure to this stuff has diminished drastically over the last couple of years because I stay a mile away from anything associated with webcams but when ww.com was still up it was a regular occurrence that we had to step in and shut down feeds and alert the authorities. And that was with a published policy of going after the originators and giving full cooperation to LE.

So yes, that is happening at a non-negligible scale. In fact it is happening at a scale that is incredible (to me, at least).

> 700K perps world-wide and an (obviously) smaller number of producers

The figures I'd want are not consumers or producers but numbers of children. A child being abused 1000 times is obviously awful but its not as bad as 1000 being abused once.

> A child being abused 1000 times is obviously awful but its not as bad as 1000 being abused once.

Obvious to you. The problem with this sort of reasoning is that children are not mathematical objects. It is also the sort of reasoning that ultimately leads to the rest of society deciding that us tech people are incapable of setting things up in such a way that abuse potential is minimized. Then you get reactive stuff like this article.

Thanks for your continued replies and balanced counterpoints to various statements (including my own).

Maybe I've gone the wrong direction, but I'm almost convinced now that it may actually be worse to make public statements linking encryption to child abuse because it potentially paints backdoors as a solution in the minds of the non-technical folks, and potentially makes child abuse closer to appearing to be a "solved problem" and distracts further from progress towards a real solution.

I'm aware now (thanks to you and some others who have horrid coalface experience) that the digital realm is indeed more of a problem than I'd considered. I can't bring myself to accept, however, that mandated encryption backdoors are worth the attention they're being given by law enforcement for the purposes they're trying to align it with.

I can't accept the backdoors either. Clipper was a bad idea then and any iterations of it in the present are just as bad. Clearly that is not the most effective way to attack the child porn problem, Microsoft, for all its faults has done more to combat this than lots of other parties combined, including most nation states. Properly funding various services including the LE branches responsible for dealing with these crimes would be the way to go and that does not require any backdoors at all. The mountains of proof that are usually unearthed are strong enough for conviction, the bigger challenge is to roll up the networks but there are ways to combat that that do not require access to the payload.