Hacker News new | past | comments | ask | show | jobs | submit login
Government Surveillance and Internet Search Behavior (ssrn.com)
151 points by mo on July 27, 2014 | hide | past | web | favorite | 28 comments



What Snowden did, or more specifically what the faceless, unaccountable and un-nameable NSA directors did was what George Orwell was trying to capture in his famous novel.

The notion of thoughtcrime and crimethink being the punishable event in the minds of the people. Where people are afraid to research something because of the perception that the secret unaccountable police will come take you away if you give the idea too much thought. And there is nothing you or anyone can do about it because the entity doing the enforcing is completely hidden. Even asking for the names of the directors and writing about what they've done is crimethink. Thinking about or asking for the document describing which thoughts are crimethink is also crimethink.

But perhaps I'm looking at this the wrong way, maybe a utopian civilization would consider thoughts of evil, thoughts of crime and intentions to harm others as a justifiably punishable event. If your neighbor is thinking about how to make a bomb, or how to kill someone, or how to commit suicide, or how to defraud and deceive others, wouldn't it be better if the secret police put a stop to it there?

We could live in a post-crime society. Where everyone is un-corruptable and all humans treat each other as lovingly as we treat our own bodies. The problem with this is that the secret police only enforce the rules of the rulers, which has a thicker script for the lower classes than the upper classes.


>> But perhaps I'm looking at this the wrong way...

Perhaps in such a civilization George Orwell's 1984 would be illegal too.

>> We could live in a post-crime society...

There is no such thing as a post-crime society. Crime is subjective. It's fine when you think about murder etc as a crime in a democratic setting where citizenry has equal rights. But, the definition of crime is never a constant - It was a crime to consume/buy/sell alcohol or drugs once, it isn't anymore (some drugs in some places). Human torture was legal once, it isn't anymore. Even now, in some places of the world, education/equal rights for women is a crime, freedom of religion is a crime. So, there is no such thing as a post-crime society. All that can be is transparency, which democracy and free press have best provided. Hidden policing/courts etc are a step backward as there is no oversight on their operation.


Perhaps in such a civilization George Orwell's 1984 would be illegal too.

Right now in Thailand: http://blogs.indiewire.com/anthony/thailand-bans-screening-o...

Also posters like http://pratyeka.org/thai-social-networking-advice.png blatantly advise people not to discuss, share or upvote dissenting views, even online.

Orwell and Zamyatin were right.


> But perhaps I'm looking at this the wrong way, maybe a utopian civilization would consider thoughts of evil, thoughts of crime and intentions to harm others as a justifiably punishable event. If your neighbor is thinking about how to make a bomb, or how to kill someone, or how to commit suicide, or how to defraud and deceive others, wouldn't it be better if the secret police put a stop to it there? > We could live in a post-crime society. Where everyone is un-corruptable and all humans treat each other as lovingly as we treat our own bodies. The problem with this is that the secret police only enforce the rules of the rulers, which has a thicker script for the lower classes than the upper classes.

In the past, philosophers have been discussing about "The problem of evil"[0] - which is the paradox of an omnibenevolent, omniscient, and omnipotent God allowing (morally) evil to exist. Obviously, it's normally discussed in a theological context, but one argument would apply to a non-theological context as well: the defense of free will. Since free will is the more important/ preferable than non-existent of evil, you can't have human with free will but without the choice of being evil. So as long as free will exists, you will have to be content with having evil/ crime etc as well.

Just to be clear, "evil" in the previous paragraph is being used in the sense of morally bad, and not the invisible, flying, human-haunting type.

[0]: http://en.wikipedia.org/wiki/Problem_of_evil


It is not clear that notion of free will is even a consistent concept however. E.g. free will is perfectly explained as what a deterministic decision process feels like from the inside.


> maybe a utopian civilization would consider thoughts of evil, thoughts of crime and intentions to harm others as a justifiably punishable event.

Can't tell if you're being sarcastic or not but you're certainly not describing my utopia. So there, the idea that an "objective" utopia exists is disproven. Taking that argument further, who now decides whose utopia becomes the utopia? I admire Snowden a great deal for both his courage and his intellect, and one of the strongest arguments he makes against your utopia (again, I just can't tell if you're being sarcastic) is to remind us of the ancient-seeming idea of a "security state". That is, close to 100% security can be had, at the cost of liberty. In other words, it's a tradeoff. We can quibble about whether that tradeoff could be improved (the equivalent of a free lunch or "Pareto improvement" in economics), but "thoughtcrime" and "crimethink" is so far beyond where I would draw the line, you can't even see the line anymore from thereā€¦


How would you know if you had a utopian society yet?

Who gets to define what is a "thought of evil"? Is freeing slaves evil, or against the natural order of things? What about considering interracial marriage? Or living an openly homosexual lifestyle?

Would it be better if the police put a stop to it before such people started doing such horrible things? Locked up people before they had a chance to make their morally corrupting case to society? Suppressed discussion so that ideas could not be thrashed out, improved, built upon and weaponised in a memetic fashion to spread like some awful disease through the minds of right-thinking folks?

We may think we're incredibly moral, because we've got to the point where we now recognise these past beliefs as wrong, even if we're still a few steps from getting everyone to accept them. But for those growing up now who realise the importance of LGBT rights, the generation before you were convinced of their morality for recognising mixed-marriage rights over their predecessors, who were convinced of their morality for recognising the evil of slavery over their predecessors.

If history is any guide at all, it's likely there are things which you think are just the natural order, that will in the future be successfully argued to be morally reprehensible. Our descendants will wonder how we ever let ourselves do such things. I've no idea what those things will be; it might be environmental destruction, or carnivorism, laws/mores against public nudity. But I'm pretty sure we want those thinkers out there, as we're no where near utopia yet.


> But perhaps I'm looking at this the wrong way, maybe a utopian civilization would consider thoughts of evil, thoughts of crime and intentions to harm others as a justifiably punishable event.

Maybe. And yet, I seem to remember reading an essay a while ago about the difference between cities and companies, and why some cities have lasted for thousands of years, whereas most companies are lucky to make twenty. The answer in the author's opinion was a tolerance for dissent.

So, you may be right about stopping violence, but do you think those in charge of surveillance will stop there? Because I do wonder whether, in a world where thoughtcrime is a real possibility, things might not get end up in a permanently rigid status quo, rather than the free-flowing and ever-changing conversation that we call culture.


Strangely the creeping Orwellian scenario may be the most optimistic view to take. If the monitoring is unnecessary it can be rolled back one day. The threat can be removed.

But what if the monitoring really is vitally important in preventing violent crime? The government will loose the spying capability soon enough as technology and understanding evolves. What will happen then?


> But what if the monitoring really is vitally important in preventing violent crime? The government will loose the spying capability soon enough as technology and understanding evolves. What will happen then?

We can look at data in regions where monitoring is most egregious and see whether improvements correlate with monitoring, or if they could be explained by other factors.


Civilization has gotten along fine for thousands of years without all-pervasive electronic surveillance.


In R.A. Lafferty's Hugo & Nebula nominated "Past Master", a time machine is used to retrieve Sir Thomas More to help fix their attempted utopian society, http://irosf.com/q/zine/article/10456

"Thomas's initial reaction to Astrobe is perplexity: how can the people of Cathead reject happiness when they have been offered it on a platter? Why choose misery over a golden life? The people of the slums inform Thomas that a miserable life is better than no life at all. As he asks questions and explores this brave new world, Thomas discovers that the best people on the planet, those of finest intellect and judgment, have migrated to Cathead."



"Using panel data, our result suggest that cross-nationally, users were less likely to search using search terms that they believed might get them in trouble with the U. S. government."

Since their data is based on Google Trends, the only conclusion they can validly draw from their research is that users were less likely to search for such terms on Google. People with sensitive search topics may have moved to a search engine like DuckDuckGo, which doesn't log users' searches.


I doubt it changed the search behavior of "Joe User" at all.


What's more important than changing the behavior of "Joe User" is changing the behavior of "Joe IT", who makes the tools "Joe User" uses. Adding end-to-end encryption to highly popular services, making it very easy to use, or even making it all work in the background without "Joe User" even knowing about it, is much more effective and faster than teaching every single Joe Use to use end-to-end encryption.

Imagine if Hangouts and Whatsapp added end-to-end encryption to their services, the way TextSecure works. BOOM - 1 billion people now using end to end encryption, most without even realizing it. That's as "world changing" as it can get, and hopefully, something like this will eventually happen on popular platforms.

Changing the behavior of "Joe Tech Evangelist" is also extremely important. "Joe Tech Evangelist" is basically what caused Chrome to go from 0 percent market share to 50 percent market share (in some stats). Google's advertising of Chrome didn't do nearly as much as every tech guy out there pushing Chrome onto their families and friends, who were still using IE, and some even Firefox.

And if "Joe IT" doesn't make the kind of services that offer really strong encryption and security, he should be very worried "Joe Tech Evangelist" will push users towards the platforms that have it anyway, and (eventually) lose its userbase. But it obviously won't happen overnight, and it also depends on the commitment of "Joe Tech Evangelist" to do this.


It propably at least changed the search behavior of the Bad Guys.


Stupid bad guys. Smart ones already assumed the NSA was poring through everything.


Take a look at the average violent criminal. The stupid ones are just as dangerous as the smart ones, and there's a lot more of them.


Terrorists aren't much like criminals. For a start, they're ideologically motivated. And that's the argument for why we as a society, need to treat terrorists differently than criminals in the first place.

There's only a very, very few terrorists, even when the US government has done its level best to create as many as possible. I doubt we can take a quantitative argument like yours and get decent results.


The term "bad guys" is much broader than just terrorist. Feel free to substitute "drug smugglers", "human traffickers", "members of organized crime groups", "militant extremists", or, yes, even "terrorists". Nor did I make any argument as to how to handle any of them. The point is that you don't have to be smart to cause harm. The smart ones were presumably harder to track beforehand; the not-so-smart ones are likely a lot smarter now.


Of course. As well as there will still be some that are even more stupid so we can have nice statistics about how successful the new tools are.


How about "how the Snowden relevations changed the behaviour with online porn"?

Because I'm considering leaving all that behind. Some creepy organisation going all peeping tom on me? No thanks.

Sure this has been researched as well?


actually the study looks at other searches which would get them in trouble with friends, neighbours and colleagues. So yes it has been researched, partly.


The quality of writing in this paper is quite terrible. I'm not sure what kind of review this went though, but I'm kind of shocked it got approved with data points named "After Prism Revelations". Revelations is an emotional word for that data descriptor.


Your comment is extremely bizarre. The paper seems exceptionally clear and well-written! It's rather short. It gives the necessary background and states its assumptions, it is very easy reading.

Can you quote a paragraph you didn't like? Is your main issue that the data descriptor is defined elsewhere (as opposed to a date range)? I don't think the descriptor itself is emotional at all.

By the way, I thought based on the abstract that the effect would be rather large. For anyone who downloads the PDF, go to the graphs - the effect is actually a very small difference even on search terms rated to get people in trouble with their governmetn, as well as on personal-privacy related search terms.

The effect is clearly there but rather small compared to what I was expecting.

Granted I only spent a few minutes with the PDF and could be misinterpreting.


"Revelations" is not emotional, it's a form of "reveal".

"After PRISM Revelations" simply means "After PRISM was revealed", since it was hidden from the public before. So, Snowden revealed the existence and scope of PRISM to the group of people whose search behavior this study is analyzing (the general public).

Not an emotional word and, overall, not bad writing in the study.


This wasn't approved by anyone in particular. SSRN is a preprint archive, like arXiv.org, where anyone can upload their work. Some of the work published there has also published in peer-reviewed venues, some hasn't. This particular paper hasn't been; the 2nd author's CV lists it as one of four "work-in-progress" drafts she currently has circulating for comments, before presumably revising it, and submitting it to a journal sometime in the future.




Registration is open for Startup School 2019. Classes start July 22nd.

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

Search: