Shostack + Friends Blog Archive


The emergent chaos of fingerprinting at airports

HONG KONG (Reuters) – A Singapore cancer patient was held for four hours by immigration officials in the United States when they could not detect his fingerprints — which had apparently disappeared because of a drug he was taking.

The incident, highlighted in the Annals of Oncology, was reported by the patient’s doctor, Tan Eng Huat, who advised cancer patients taking this drug to carry a doctor’s letter when traveling to the United States. (“Cancer patient held at airport for missing fingerprint“, Reuters, May 27 2009)

Reuters classifies this as “oddlyEnoughNews,” but in fact it’s not odd at all that over time, additional layers of “no” will expose conditions unimagined by their designers. Chaos will emerge. In a free society, that chaos is an accepted part of life. We stop only that which is explicitly denied, not that which the designer didn’t anticipate. In information security, we often default to deny, because we know our imaginations are limited. But the role of security in society used to be carefully limited, for precisely these reasons.

(Via Slashdot)

4 comments on "The emergent chaos of fingerprinting at airports"

  • Dean Loomis says:

    Why do they want your fingerprints? Because they think they can use them to distinguish you from other people. But they dare not actually test if this is true.
    Elementary molecular studies have shown that Africans have the greatest genetic diversity while Asians are much more genetically uniform, while “Caucasians” range somewhere in the middle. Is this lack of diversity also apparent in fingerprints? One or two studies with small samples have suggested that this may be true.
    But the studies that security agencies rely on to prove that fingerprints can actually distinguish among individuals have been based on populations that are not controlled for genetic or even “ethnic” diversity. They use populations of U.S. criminals, or Englishmen, or even whoever they could find in the building where the biometric product was developed. Attempts to normalize against any defined target population are exceedingly rare, and normalization against the global population needed for a transportation security agency charged with protecting international borders, or needed by the security group of a multinational corporation, are nonexistent as far as I’ve been able to discover.
    The math is similar to the math used for cryptographically analyzing the likelihood of collisions in hash algorithms, with the developmental process that leads from each embryo’s genetic endowment to the fully differentiated fingerprints that each of use are born with functioning as the hashing algorithm. If the algorithm doesn’t produce very much randomization, as the data suggest, how many runs does it take before collisions occur and erroneous matches between fingerprint signatures and identities will occur, leading to arrests of innocent people for traveling with improper credentials?

  • enlargement says:

    Very much considerate post but there are some place where I will not agree. But overall its altogether good.

  • Great headline. If your cookie has a bite-sized action and your reader completes the action, I think two things happen. Their self-confidence goes up (which feels good) and their trust in you increases.

  • nepHaptealt says:

    Hey there everybody i know you’ve all probably heard this already but check out google for all the scandal thats been going around about her lately
    basically she first claimed that the tape was for a boyfrind in highschool at which time she begged him to tell the media she was 17 at the time. This morning on tmz they talked about a voice interview where the man who had the video insists that she was 20 at the time.
    You can check out the newest tmz story and pictures here.
    So i know theres a lot of bullshit going around on the internet but i finally found the actual video. SO CHECK IT OUT SHES HOT AS HELL but all in all just another homeade sex video. The link is

Comments are closed.