Got linked to this by @willsmith. Thought it was fascinating enough to save a copy of my own.
If we ask whether a fact about a person identifies that person, it turns out that the answer isn’t simply yes or no. If all I know about a person is their ZIP code, I don’t know who they are. If all I know is their date of birth, I don’t know who they are. If all I know is their gender, I don’t know who they are. But it turns out that if I know these three things about a person, I could probably deduce their identity! Each of the facts is partially identifying.
There is a mathematical quantity which allows us to measure how close a fact comes to revealing somebody’s identity uniquely. That quantity is called entropy, and it’s often measured in bits. Intuitively you can think of entropy being generalization of the number of different possibilities there are for a random variable: if there are two possibilities, there is 1 bit of entropy; if there are four possibilities, there are 2 bits of entropy, etc. Adding one more bit of entropy doubles the number of possibilities.
Because there are around 7 billion humans on the planet, the identity of a random, unknown person contains just under 33 bits of entropy (two to the power of 33 is 8 billion). When we learn a new fact about a person, that fact reduces the entropy of their identity by a certain amount. There is a formula to say how much:
Î”S = – log2 Pr(X=x)
Starsign: Î”S = – log2 Pr(STARSIGN=capricorn) = – log2 (1/12) = 3.58 bits of information
Birthday: Î”S = – log2 Pr(DOB=2nd of January) = -log2 (1/365) = 8.51 bits of information
Note that if you combine several facts together, you might not learn anything new; for instance, telling me someone’s starsign doesn’t tell me anything new if I already knew their birthday.
In the examples above, each starsign and birthday was assumed to be equally likely. The calculation can also be applied to facts which have non-uniform likelihoods. For instance, the likelihood that an unknown person’s ZIP code is 90210 (Beverley Hills, California) is different to the likelihood that their ZIP code would be 40203 (part of Louisville, Kentucky). As of 2007, there were 21,733 people living in the 90210 area, only 452 in 40203, and around 6.625 billion on the planet.
Knowing my ZIP code is 90210: Î”S = – log2 (21,733/6,625,000,000) = 18.21 bits
Knowing my ZIP code is 40203: Î”S = – log2 (452/6,625,000,000) = 23.81 bits
Knowing that I live in Moscow: Î”S = -log2 (10524400/6,625,000,000) = 9.30 bits
How much entropy is needed to identify someone?
As of 2007, identifying someone from the entire population of the planet required:
S = log2 (1/6625000000) = 32.6 bits of information.
Conservatively, we can round that up to 33 bits.
So for instance, if we know someone’s birthday, and we know their ZIP code is 40203, we have 8.51 + 23.81 = 32.32 bits; that’s almost, but perhaps not quite, enough to know who they are: there might be a couple of people who share those characteristics. Add in their gender, that’s 33.32 bits, and we can probably say exactly who the person is.
An Application To Web Browsers
Now, how would this paradigm apply to web browsers? It turns out that, in addition to the commonly discussed “identifying” characteristics of web browsers, like IP addresses and tracking cookies, there are more subtle differences between browsers that can be used to tell them apart.
One significant example is the User-Agent string, which contains the name, operating system and precise version number of the browser, and which is sent every web server you visit. A typical User Agent string looks something like this:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:220.127.116.11) Gecko/20070725 Firefox/18.104.22.168
As you can see, there’s quite a lot of “stuff” in there. It turns out that that “stuff” is quite useful for telling different people apart on the net. In another post, we report that on average, User Agent strings contain about 10.5 bits of identifying information, meaning that if you pick a random person’s browser, only one in 1,500 other Internet users will share their User Agent string.
EFF’s Panopticlick project is a privacy research effort to measure how much identifying information is being conveyed by other browser characteristics. Visit Panopticlick to see how identifying your browser is, and to help us in our research.