Time for Internet Engineers to Fight Back Against the “Surveillance Internet”
technologyreview.com
technologyreview.com
It is going to be the same patttern with all forms of electronic communication. The spectre of weapons of mass destruction (anything that can kill more than one person at a time) is the justification/motivation for government and law enforcement to monitor everything online. Whether it has merit is kind of besides the point. It is an advantage, perceived or otherwise, that is not going to be given up, under any circumstances.
Another part to this is whether conspiring online actually constitutes a crime. People can chat all day about terrorist plots but until someone actually goes out and starts building a bomb or procuring weapons has a crime really taken place? Only when a society can decide on the answer to this question will it be possible to address online surveillance.
The "big picture" for me are the effects of the loss of privacy. What was once one of the most horrendous forms of punishment/torture, is now mainstream. Also, this only gives more power to already too powerful of people (you already mentioned the nukes). I think we've already proven we don't need mass surveillance to not blow ourselves up.
> Another part to this is whether conspiring online actually constitutes a crime.
""To be classified as a crime, the act of doing something bad (actus reus) must be usually accompanied by the intention to do something bad (mens rea), with certain exceptions (strict liability)."" - http://en.wikipedia.org/wiki/Crime
In other words, it's up to the most powerful, who are also in control of the surveillance to decide this, not us. Also, the "big picture" issues here aren't really surveillance itself, but MASS surveillance, SECRET courts, and CLASSIFIED evidence.
Everyone's intention? In other words everybody is the opponent. Is that the big picture you have about the world?
I would give you another +1 for mentioning the AGPL, if I could.
I mean, if the NSA/CIA/FBI can get away with this kind of surveillance, and seeing the scope of "Tailored Access", why wouldn't they have rigged the last few elections, too? Clearly the security around voting machines isn't nearly as tight as that of Level 3,
http://www.theatlantic.com/politics/archive/2013/11/the-surv...
https://www.noisebridge.net/wiki/Worldwide_Aaron_Swartz_Memo...
I don't see a very good future of fighting the 'surveillance internet' if we don't face the notion that once data is out there -- of lots of people, in one place -- NSA will always be interested in it and will probably get it. Pragmatically speaking, pulling out of Facebook, Google, etc. is the only surefire way to fight the surveillance internet.
So no, leaving Facebook or Google solves only a tiny part of the problem.
I also agree that you could make a convincing case that aside from the legality and broad social unacceptability of the collection, the way data is currently used by the government is much much more in line with the surveilled's interests than many of the businesses gobbling up consumer data. Corporate surveillance isn't inherently any better than government surveillance. Pretending otherwise makes us enablers.
Can Facebook maybe deny me a business license? Cancel my passport? Is there a Facebook commando that would break down my doors if I did something Mark Zuckerberg didn't like? No?
Can we then stop pretending like corporations and governments play even remotely in the same league?
What can Google actually do? Target me with ads? I won't so much as see them.
And I don't use facebook, it barely raises an eyebrow. I couldn't try that with a national ID because I wouldn't be able to do anything.
Give everything to the NSA, IRS or anyone else that asks.
Edit: this one https://news.ycombinator.com/item?id=6680763
A few months ago they were using the "metadata" argument. Where are the prosecutions now that thats obviously BS?!
Yep, it's absurd to the extreme to compare the NSA and Feds to Google.
Google, Facebook, et. al. can willfully share your information with the US or any other government.
AT&T gets paid $10 million annually for access to your data, and as a publicly traded company, if they don't do that, they're not being fair to their shareholders, to whom they've promised to maximize profits.
http://articles.washingtonpost.com/2013-07-21/world/40713603...
The reverse is not true, so if only the government is surveilling you, you can still stay out of some of the corporate databases.
They did it with boeing and airbus http://www.economist.com/node/1842124 and who knows how many other companies.
And who knows what boeing did in return to gain the favor of NSA...
Articles like these presume that it is OK for the government to access whatever it can technically get its hands on, in spite of Constitutional constraints and laws that do or should state otherwise.
When the NSA approaches Google, I want Google's General Counsel to deny the request with solid legal standing. Likewise with backbone providers and on down the line.
We are either a nation of laws or we're not. Our government is either beholden to those laws or they are not. Had we the proper laws and commitment to our Constitution, then Snowden's revelations would have resulted in trials and prosecutions. Instead, too many seem to be ceding to the government the right to surveil its citizens with impunity, and are instead focusing on technical defenses against their own government.
It's lunacy, and if the primary emphasis is not on legal redress, then we have already lost.
The NSA is a rogue agency that does not respect laws (or reinterprets them as they see fit). Going through the legal process to shut it down is certainly worthwhile, as is throwing its criminal elements in jail, particularly those that are happy to lie in congress.
However, the reality is that a rogue agency can evolve in the dark corners of the government, and that therefore it is likely that it will happen again. And even if it never happens in the US again, there are other countries out there, you know?
A strong technological solution that makes large-scale snooping impractical is a sine-qua-non no matter what happens on the legal side.
I got the news. They approach Google AND they plug into private lines. The latter case is what I referred to when I mentioned "backbone" providers. Again, I want any private entity to have legal standing to refuse NSA requests.
>The NSA is a rogue agency that does not respect laws (or reinterprets them as they see fit)
I agree that if an agency goes rogue, then laws are only retroactive. That is, laws provide a penalty that is triggered only after an offense has occurred. But, clear (i.e. not ambiguous) laws with clear penalties can be a powerful deterrent. Whistle-blowers like Snowden are then empowered to stop abuses and illegal activity. They are automatically branded as heroes instead of traitors who must flee the country or worry for their safety. As it is, the good guys like Snowden are being put on the wrong side of the law and vice-versa. This must change.
>However, the reality is that a rogue agency can evolve in the dark corners of the government, and that therefore it is likely that it will happen again.
That's true and always has been. But, we don't just say "well, laws will be broken, so let's not bother having them". It's really the entire point: to prescribe what is acceptable behavior and provide penalties for violations.
>A strong technological solution that makes large-scale snooping impractical is a sine-qua-non no matter what happens on the legal side.
We actually agree to some extent. I don't advocate that we not implement technical measures. Where we depart is on priority. The wording of your last sentence signals this departure. I would flip "technical solution" with "legal side".
Ultimately, if the emphasis is on technical solutions, then we will all be pwned with impunity. Period. Are you going to write your own firmware? Manufacture your own chips? Are you going to personally write all of the security and other endpoint software in your stack, including the OS? Even if you did, would you be able to guarantee zero vulnerabilities in your own code?
Checking rogue agencies, providing more oversight and enforcing clear laws are the only way out. Technological solutions are but a backstop that we hope will provide us with some defense in the event that a rogue agency goes undetected for some period.
> I got the news. They approach Google AND they plug into private lines. The latter case is what I referred to when I mentioned "backbone" providers.
As far as I understand, these were not lines provided by "backbone providers". These were private lines laid and paid for and owned by Google. There was no third party who bent - Google got pwned directly, in secret, with impunity.
"[W]hy don't we also make the NSA's activities illegal and/or enforce existing laws which already state as much?"
Who is "we" in your proposal? The people of the society? Some percentage don't mind the snooping - they think it is protecting us from terrorism. Another percentage object, and a third subset have little idea of what's going on.
Maybe you imagine there is some approximation of democracy in the US. Why do you think so? How well do you think the legislators represent the interests of citizens? How many viable parties compete for votes? Can you even verify that the announced election results have any definite relation to the votes cast at electronic machines?
Some elements in Congress have tried regulating the surveillance apparatus, at least since the Church committee in the 1970s. Obviously the effort has not succeeded, and I'm not perceiving much prospect of improvement on that front.
The only realistic avenue to improving privacy and digital security is some combination of (a) securing endpoint systems (b) an open-standard, no-back-door solution for pervasive encryption of data in transit and (c) services that somehow offer secure communications hubs without the operators having to shut down, sell out or be imprisoned.
These guys are working on (b), and I applaud their efforts.
By "OK", I don't mean that you agree with it. I mean that by assuming the posture that you are going to focus on defending yourself technically, you are effectively ceding that the government has the right to access whatever it can. You're telling the government that it's OK to get whatever you are unable to protect. It is absolutely the wrong posture and it's what the NSA wants.
>Who is "we" in your proposal? The people of the society?
Yes. Exactly. Who else would "we" be? What does it matter that some don't get it yet when there is some group that finds it objectionable and contrary to our fundamental rights? I mean, it's really a strange position to take when you think of it. If the government was coming after you with tanks and guns, would you go buy a shotgun and stronger locks for your doors? Or would you spread the word and work to check the government per the laws of the land?
So, how about a grassroots campaign to change the public perception landscape?
>Maybe you imagine there is some approximation of democracy in the US.
An approximation of democracy is exactly what I think we have. Listen, you're preaching to the choir on the whole "our democracy is broken" thing. But, to the extent that there's anything left or any way back, fixing this democracy is our only hope. Again: grassroots.
It boggles my mind that the response is to cede everything--broken democracy, surveillance state, etc.--then just dig in to prepare to defend oneself from his/her government. If you actually believe that you can defend yourself from government lawlessness, then you ought to believe that you can play a role in fixing the government.
>Some elements in Congress have tried regulating the surveillance apparatus...I'm not perceiving much prospect of improvement on that front.
And, how long have we been patching zero-day exploits, fending off viruses, etc.? Why are you so confident that you are capable of defending yourself from a determined government with unlimited resources?
>The only realistic avenue to improving privacy and digital security is some combination of [technical solutions]".
I am not advocating that we don't take prudent, precautionary technical measures. Of course we should. I'm saying that they should be secondary and we must lead with legal measures to have any hope of reclaiming our rights and putting an end to this. All of the things you listed sound great. But, in truth, we know that there are exploits and vulnerabilities everywhere. Ironically, as technologists, we want to believe in technical solutions. At the same time, we also know better than anyone that a determined adversary will find a way to pwn us.
For instance, all of your technical "solutions" rely on an up-to-date OS with no zero-day vulnerability, perfect non-exploitable endpoint software, and even full knowledge of our hardware stack, firmware, etc.
In general, what you propose is exactly what the NSA would want. And, you will lose. If you are fortunate enough to even know when you've been pwned, you'll patch things up until the next time. You'll just play cat-and-mouse while your information floats away. Then, at some point, you will say "Wait, this is what we normally do to defend ourselves against criminals. Why am I having to do this to defend against my own government?"
At that point, you will realize that this is first and foremost a legal problem. Only then will you demand what I am advocating right now.
That being said, I hope IETF starts working on a new highly secure Transport layer protocol to replace TCP, within the next 5 years, and I hope they use Dan Bernstein's CurveCP [1] for inspiration.
We need the Internet encrypted and secure by default, and I don't care what Google or other advertising companies have to say about it. Adapt or die. Security of the web and the protection of the human right to privacy is way, way more important in my book. If they choose to fight such a move, instead of adapting and actually supporting it, then they will have become the enemy, and they'll end up on the wrong side of history.
So IETF's goal should be to get everyone to switch to these more secure, already existing protocols, and implement them within a year, or two at most.
In the meantime work on replacing TCP within the next 5 years, and also think about ways to create a new secure-by-default and easy to implement, IP-level protocol, to be used within 10-15 years.
If we are to "take the Internet back", then it needs to stop being such an easy tool for mass surveillance, so in a way, we need to replace all of its insecure parts.
[1] - http://curvecp.org/