Take the 2-minute tour ×
Information Security Stack Exchange is a question and answer site for Information security professionals. It's 100% free, no registration required.

As Google Glass slowly starts shipping, I've recently came across an article on stopthecyborgs.org that mirrors my anxieties that Google Glass currently does not feature a robots.txt for my face. It is absurd that while I can protect my website's privacy, as it currently stands, I cannot do this with my face anymore.

Personally, I would like any attempt-to-recognize-my-face to be an opt-in feature, not an opt-out feature. From a privacy stand-point, the glass should only transmit any data about people if it recognizes an 'OK you can post me' sign on them, before sending any transmission. It poses a much greater security risk if it is an opt-out feature than if it's opt-in. But currently, there are no provisions at all, for either opt-in or opt-out, that I know of.

If Google Glass gets widely adopted, this effectively opens up the possibility for a total corporate surveillance (and if CISPA and similar attempts are not stopped, possibly also opens up to US governmental surveillance).

As some people resist the idea that eliminating privacy should be a good thing, the question is if a robots.txt functionality can be implemented that protects privacy against the glass and it's future, more miniature, competitors.

As some countries are already trying to enforce policies to forbid to hide your face with a scarf, this becomes a problematic task to do on your own, especially in summer.

Are there any attempts to research if people are aware of or troubled by this threat? Does Google have a public stand-point over this matter? Has anyone studied or proposed potential technical (or other) solutions to this problem?

share|improve this question
3  
While I do like this question, I don't think it belongs here. It's very open-ended and it's simply unanswerable. –  Adnan Apr 20 '13 at 9:25
1  
@Adnan I share your concern and if you could provide some details, I made this meta question in order to identify what makes it open-ended and how improve the question meta.security.stackexchange.com/questions/1237/… –  naxa Apr 20 '13 at 9:37
1  
You asked this very oddly, but not entirely in a bad way –  David Apr 20 '13 at 23:35
    
"As an added precaution, Superman would vibrate his face slightly so that photographs would only show his features as a blur" - Maybe this is an option to consider? :P –  Bart van Heukelom Apr 22 '13 at 18:20
    
superheroes against facial recognition (german) <a href="de.slideshare.net/vadimn/…; title="example">Slideshare Link</a> Not only CV Dazzle, also to spam and to tag yourself wrong will help in the long run. –  user25170 Apr 25 '13 at 7:21

2 Answers 2

up vote 7 down vote accepted

There is a fascinating technology that allegedly works to block recognition of a human face through use of computer software created for that purpose. The project is called CV Dazzle, "Camouflage From Computer Vision" by Adam Harvey ([main site])2.

If you Google CV Dazzle, you will find dozens of additional links to material at the main website and in reviews of the technology.

share|improve this answer
1  
While this research is interesting, and a +1 for the links, I'm afraid it's not solving the problem of remaining anonymous at all. Identification based on our biometric properties has progressed way past facial recognition alone (not that it was ever highly reliable). A lot of research has already been done in walk detection where facial recognition wouldn't be reliable or possible. Other symmetry patterns have also been suggested, e.g. patterns in movement of our heads, among many others. –  TildalWave Apr 23 '13 at 20:29
    

Although Google Glass makes the issue look whole new and technological, this is actually very old news. To make it short, there is no privacy in public places.

Indeed, whenever you go out on the street, you show your face to the World at large, opening up the possibility that any bystander recognizes you. This is by design. That's how civilization in general works. To make it more precise, individuals are, by nature, able to conceive their own plans and act on their own (some people would call it free will). In a society, people must cooperate to handle shared resources, be it food, water, or simple space. Laws define the boundary of free actions, but enforcing laws is expensive; there just are not enough law enforcement agents to verify that everybody acts lawfully (including the agents themselves). Society can remain lawful only because most people are basically honest and law-abiding. Why are most people honest ? Mostly because they were educated that way, and kept under, let's say, continuous formation, by social pressure, which in turns relies heavily on non-anonymity.

It is depressing, but well recognized, that when people feel that they are anonymous, many of them will begin to act as jerks (not all of them, perhaps not even most of them, but sufficiently many to become a big issue). In pre-urban societies, there is no privacy. For instance, consider how indigenous North Americans lived before the arrival of Europeans: they lodged in longhouses, which are communal buildings, meaning that in order to be alone, they had to walk away from the village. Later, when housing could provide some one-person rooms, it became possible to have a bit of privacy -- but not much. In an 18th century New England village, everybody knew everybody, and the people as a whole felt entitled to have strong moral opinions on what everybody else does at all times. Gossip was not only tolerated, but actually encouraged, in that it allowed moral pressure to be exerted.

Later times have brought the luxury of privacy, mostly because of the sheer size of cities: it is no longer possible to know everybody and be known by everybody, so you keep on meeting total strangers who you do not recognize, and who do not recognize you. Space is separated between public space, morally shared by everybody, and private space, claimed to be "his own" by some individual. Laws theoretically apply to everybody regardless of where they are, but some laws are specific to the public space. This includes, of course, laws about decent clothing (you can be naked in your home, not in the street), but also, and that's where I was going, about concealing your identity.

A "robots.txt for your face" is called a mask. That's hardly new technology. And there are laws about masks. That page lists laws in various US states; many other countries in the World have similar regulations, and they got some revival recently, in some European countries which passed bans against integral Muslim veils in public areas (e.g. in Belgium). Although these recent laws were fueled by some controversial undertones (some people are against Islam in general; some people view it as a tool to emancipate women;...), the official and theoretical core of these laws is the basic principle which I have alluded to above: in public areas, you shall not be anonymous (potentially). This means that while there is no obligation to shout your name, address and blood type every ten seconds, it should still be possible for bystanders to see your face and be able to recognize you later on, should an inquest about some crimes be run. Late actuality in Boston has shown this principle at work: in times of crises, witnesses (and amateur videos and photos) can be used to identify suspects.

Whether all of this is morally good or not is an open debate; but, right now, continued operation of society relies on this potential non-anonymity of citizens when in public areas. Changing that is akin to people all wearing masks when going outside. This is a revolution of all social structures. Controlled experiments of that kind have been performed for centuries; these are called carnivals: valves to evacuate social pressure overflows, under strong boundaries to avoid the whole thing degenerating into chaos. It is conceivable that a society with strong anonymity in public places is possible, but it would still be a far-reaching change from today's state of affairs.


The Internet brings on a twist on all this. Indeed, on the Internet, people's identity works as a whitelist, not a blacklist. A mask is a blacklist: it blocks some specific features of your identity. On the other hand, on the Internet you show only that which you are willing to show. One billion Facebook users are apparently willing to show a lot (and, to their own surprise, a lot more than what they actually expected), but it is still mostly a conscious push effort.

This yields the pervasive feeling that Internet provides anonymity by default. This apparent automatic privacy, and the lack of biological feedback when you interact with other people over the Internet (typed text is not as rich as voice modulation and facial expressions), tends do disinhibit people. A cursory look at comments on Youtube will show that the Internet seems full of jerks... which is an illustration of the possible clash between "civilized society" and "anonymity". Being anonymous is not bad in itself, but it opens the possibility for people to show how much disrespectful they are, and, for some of them, that "how much" can reach high summits.

When you mix the Internet and social behaviour, you get interesting clashes. Disparaging comments in public forums, abrupt discovery of the lack of assumed privacy of your Facebook data, and Google Glass, are three facets of that kind of clash. Google Glass is not qualitatively different from hiring a private detective, to identify every people you encounter in the street; but it is much cheaper to the point that it makes it possible even in big urban societies. This may force us to reconsider some social rules.

For instance, some people (like you, apparently) react to the possibilities offered by Google Glass by looking for a mask -- which laws do not permit (depending on the country, of course; in Japan, they allow wearing masks, and many people do -- but they have their own ways with social pressure).

Ultimately, Google Glass brings us back to the "Village" stage, where everybody knows everybody. So it is not new; we have experience about what that means. And it is not specific to Google Glass either.

There has not been much rational effort toward individual privacy. Most of it came as an unexpected boon from the size of cities. In recent centuries, the way to escape the stifling ambiance of small villages has been to move to either isolated places where you could be all alone, or to crowded places like big cities. I don't expect it to change much. If we consider the example of social networks, we will see that most people don't really care about their privacy with regards to people they don't know. People post their photos on Facebook for the benefit of their friends and acquaintances, and are not overly worried that the whole World can see them as well. What worries people is what people they know will think about it (e.g. see the panic spread on the face of a college student when he learns that his dear mother has decided to "go on the Internet" and will thus gain access to all the photos demonstrating how the said student organizes his evening homework).

Indifference is the new "big city" for anonymity. My guess is that this is how society will react to privacy issues raised by Google Glass and similar technologies. In the (recent) past, you had privacy by virtue of nobody being able to associate your name with your face. In the future, you will have privacy by:

  1. assuming that complete strangers do not care about associating your name with your face;
  2. yourself not caring about whatever associations complete strangers can come up with.

It is up to you to decide whether you will be elated or depressed at this prospect.

share|improve this answer
5  
There is a difference between recognizing people that you saw earlier on the street and storing everyone's whereabouts in one big database, accessible to people that you cannot approve and you don't even know when anyone accesses the data. –  Luc Apr 21 '13 at 17:35
4  
It's answers like these that make me addicted to stack exchange... –  amyassin Apr 23 '13 at 14:15

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.