Over ten years ago, I used to work for a company that provided system [and device] to track when employees arrived and left the building.
Most of these employees work in the fields handling heavy equipment. Most of them had no fingerprint since they'd been worn out. We ended up using palm recognition.
The idea that "all people have a unique fingerprint" is a widely believed falsehood.
Anyone who thinks of the condition as a disease or deformity or handicap, is likely to agree that it should be "cured" if possible. After all, it's a public health harm to everyone else allowing them to breed and inflict their deformity on innocent babies that would have otherwise been normal.
And then one day an actual disease based on a germ of some sort comes along that just loves to live in the valleys of human fingerprints, and the entire human race dies because we erased the few people we had that would have been immune.
I don't understand the excuse for failing to recognize that the value of diversity outweighs it's inconvenience. I don't know how long that has been grade-school knowledge. If not multiple hundreds of years, sure at least it's been a known thing since before the brith of any currently living person?
We are turning ourselves into human Cavendish banannas.
I don't think the problem you're complaining about exists.
Eugenics have fallen out of fashion many decades ago, and most countries have laws to protect people with unusual medical conditions/disabilities - typically, the more developed they are (i.e. the more resources they're able to spend), the better the protections.
Iceland only sees 1-2 births with Down syndrome per year, and not because they've identified a way to prevent a full or partial copy of chromosome 21. They test and terminate.
My personal experience (and that of many friends) is that western OBGYNs push prenatal testing and corrective action quite aggressively. (There is reasonable room to disagree about what their motives are, and the profit motive may be sufficient for such testing.) And many people want it.
We can debate the ethical implications separately, but regardless of where one lands on that issue, eugenics has very much been normalized.
Base rate information: Iceland's health system sees about 5000 viable pregnancies per year. This means that we'd expect about 4 births with Down syndrome per year. We see 1-2, meaning that about 2-3 pregnancies are terminated in Iceland due to Down syndrome each year.
That 2% statistic is a bit misleading. Those tests carry a non-negligible risk of miscarriage, so they are generally only prescribed if other tests show that a fetus is high-risk for one of various disorders. Furthermore, some women choose not to get the test even if they are high-risk, because they would never terminate. I’m not sure it’s possible to estimate the ‘true’ termination rate using these statistics.
I've tried to find a definition of eugenics and it's very fuzzy. I feel like we could do better by adding some modifier that differentiates removing "undesirables" and terminating pregnancy which is likely to result in death or low quality life under constant care. Otherwise talking about eugenics in general becomes a bit silly. In the same way we have different words for murder and euthanasia.
Bunching those together only seems to empower religious extremists who want to prevent any kind of abortion. (see Poland recently)
Euugenics is just as ridiculous as the idea that a single disease could actually kill everyone. In both cases, sure it is actually theoretically possible, but really it was just a contrast=100% illumination of the idea not literal.
Did you think that I was seriously suggesting a disease could do that literally? Have you never encountered an idea that needed to be amplified to show it's essence?
The real life form of the process, of which eugenics is the ridiculous hyperbolic contrast=100% version, is simple pressure, discrimination, disadvantage, death by 1000 cuts, attrition etc.
Things which stick out tend to get eroded smooth. Simple as that. Not freaking literal eugenics.
Similar story here. The company I worked for provided devices which would create and read a topological map of the back someone's right hand.
There were a couple of instances where users would be missing fingers or had deformities that would throw the machine off. In these cases we would have them put their left hand upside down in the device and map that instead.
In cases where there was issues with the original and alt methods we could remove the check altogether (employ would place their hand and it would go through so nothing was suspicious). This only happened once in the hundreds of thousands of people I'd deployed to.
As an aside, this method of tracking users in and out of work was markedly more digestible than reading someone's fingerprints and increased user adoption.
Working in 1st 2nd and 3rd world we heard all kinds of stories from employees not wanting to provide fingerprints from not wanting to be framed for murder to stealing a part of your soul by extracting sacred finger marks (similar story to photos in some cultures).
It's disappointing how rigid society can be: it's my way or the highway.
When someone doesn't have a particular, not inherently required characteristic to access something, there should be a series of fallbacks, but all too often this just excludes people.
This can be seen in all areas of life, in different severities:
- the entire topic of accessibility is about letting people make use of the abilities they have despite the ones they don't have. Not having functional legs doesn't have to prevent people from filing a document in the office.
- technology which once was optional but never essential becomes required with time. Booking a doctor's visit over the phone only, buying goods over the internet only, filing taxes only with a computer, accessing a bank account only with an Android program. All that with no fallback to the post office or a personal visit.
- the requirement to pass payments through an intermediary is getting more common. Gone are the days of paying bills or getting salary directly at the office, and the sphere of buying at the point of sale is losing the cash fallback, if stories from Asia are to be believed.
None of the technological changes are bad as options, but when they become the only choices, they work to disenfranchise people, especially ones related to banking and travel freedom. Debit/credit cards are not accessible to a huge chunk of the world's population, and even in Europe, getting a bank account as an immigrant can be hard.
Another thing that strikes me in the article is calling the condition "disease", without providing any evidence that it has any negative health impact. On the other hand, it matches the definition of disability as something the society imposes [0]. That would make this a paradoxical case of someone not impaired but disabled.
Yet at the same time, that series of fallbacks has a tendency to become an attack vector on the process and a weakness in the system.
To take a tech example, TLS downgrade attacks: you want to keep things working for older clients, but it’s possible for some party in the middle to intercept the traffic and feign non-support of the newer version (or stronger cipher suites), in order to push both parties to the older version or cipher suite, and if that older version or cipher suite has anything broken about it, now you have an easier target.
Applied here, if you allow a no-fingerprints route that’s less secure, be aware that attackers will devise ways of getting onto that path for their own malicious purposes. Perhaps they’ll forge a doctor’s certificate, perhaps use some kind of acid that temporarily removes their fingerprints, that kind of thing.
It’s an unfortunate balance. With TLS, you decide eventually at some point that the problems of still supporting the old, less-secure version are too serious, and so tough luck for people stuck on the old version. With this, I certainly hope it would err much more on the side of not locking people out of the System. (Fortunately, although this particular condition is very rare, there are a much larger number of people that have damaged fingerprints or fingerprints that don’t work in whatever readers get used.)
There's no reason any other form of biometrics is less secure than fingerprints. My read of the article is that either the idea of someone not having fingerprints never came up or if that situation was considered it was filed under "we will handle the exceptions via a process to be developed later" and no process was developed. Worse, a process was developed--since we see in the article that some of these people were able to get ID cards and passports, documents requiring fingerprints--but it imposes a massive time barrier on the people.
For one-offs like obtaining an ID card, a one-time imposition of a bit more time might be reasonable. "We've installed fingerprint readers at every ___location where you can apply for an ID card but retina readers are more expensive so they are only available at the main government office" may be fine if you're referring to a passport issued every ten years. But requiring a fingerprint for a SIM card yet having no backup for that at point-of-sale is not.
This is the flaw in every single system designed by people who largely have no experience with people who have a difference in physical or mental characteristics than the designers. We see it in facial recognition trained primarily on people of a certain skin pigment. Social systems designed by people who are neurotypical. Transport systems implemented by people assuming a base level of technological or literacy skill. The list goes on.
The people doing the designing aren't being inherently mean or malicious, they're simply not thinking. And not thinking is very easy, but makes systems that are very impenetrable to people not included in that thinking.
> There's no reason any other form of biometrics is less secure than fingerprints.
Anything which is a non-standard fallback with limited use is likely to be less secure. Not because the method would be inherently worse, but just because reduced use means that less attention and effort goes into building and maintaining it.
I once watched someone crash every fingerprint reader in an extremely high traffic border crossing. The police kept asking him to try the next one, much to the dismay of everyone queueing. The machines would BSOD and they'd shut that kiosk. At the time, I joked that his fingerprints were malware or specifically designed to cause a buffer overflow somehow.
Sounds to me these men are victims of abusive software bugs. They have a finger print... it may not be unique or normal. But pressing their inked finger onto paper will leave a print. I assume the software is looking for common pieces of a fingerprint to work. Said software isn’t considering the edge case of a smooth or mostly smooth finger pad.
The fact that the government workers are letting buggy software determine eligibility for drivers licenses and passports is the real issue here. It’s really sad.
Just reading this makes me so frustrated. I can’t imagine how frustrating it would be to be them. I hope this article helps these men get the exposure they need to end this.
It's not a software bug. People did this sort of thing before fingerprints. It used to be fairly common for immigrants and refugees to not know their birthday. The beurocracy was often unable to cope with that idea, as they wanted to use name and date of birth as the unique identifier. A lot of people just made something up to make immigration happy, which then resulted in problems down the line as they didn't recall what they made up.
Indeed! As a software developer in Hong Kong, I learned that the "month of birth" and "day of birth" are optional fields - some people who fled from Communist China to Hong Kong only knew their birth year, and the ID card just shows something like "1946" instead of a full date.
I’m a woodworker. Hand Sanding always wears away my fingerprints so that I have to manually type in the password on my phone instead of using Touch ID for about a week afterwards until they grow back.
A fingerprint is there to either identify you or confirm that it is you. Now, if we would allow the software to accept a smooth fingerprint as assign it to a person then this would allow anybody to impersonate that person without additional identification means. This is not a software bug, but a security feature.
Of course the software should be able to detect that there is no fingerprint and require the operator of that software to check for the identity by other means.
Sometimes we don't always have human operators and more things are automated.
Yet another proof you need another factor of authentication. From my experience with multiple mass enrolments of fingerprint authentication of software (more than a million people combined) about 20% don't have usable fingerprints.
Let me repeat that - never rely a 100% of any piece of biometric or any other type of authentication. Have a fail-safe mode for each.
And what about people with missing fingers or hands? Or those who work manual labor and their fingerprints simply get scarred too much to match their records?
Small correction: biometric properties are a form of identification, not authentication. Identification of the body part in question, not the person that is (as elegantly demonstrated by Wesley Snipes in Demolition Man).
As always with these techniques: step one is to introduce a system that covers 99.99% of people. Step two is to let the remainder fight their way through court while getting financially ruined until a judge orders you to facilitate them too. Governments can mitigate step two by just hoping these exceptions will give up.
It would qualify as 'something you have', no? No different than a password proves you are something that knows a password, or a yubikey proving that you are someone who owns a particular yubikey. It would seem that without some yet unknown mindprinting technique everything is an identification method.
The yubikey is seeded with a secret and produces outputs that cannot otherwise be replicated without that secret. Fingerprints are hardly secrets and can be easily replicated from all of the items you touch on a regular basis.
But that also makes them poor methods of identification, no? If your fingerprint could be replicated off of a glass, than the person that possesses it may or may not be you, but they sure do have your fingerprint. Something you have!
Well if you want to get really technical, often identification works just fine, but then you get issues when you do verification, for example in authentication.
Quite a significant proportion of people have fingerprints which cannot be read[1]. Phone fingerprint readers never work. This causes no end of trouble with airport immigration in various places. I spent a couple of hours in a back room at Boston airport while various people attempted to read my fingerprints (I have them, but they never work in those airport machines).
Why do they not work though? I mean, the reader should just be reading a series of lines. Are they too close together in your case or something, or too twirly? I appreciate that this may be a personal question but I hope you can overlook this given that we are on a semi-technical forum. It seems an interesting problem (though I am not a fingerprint engineer)
You can barely see them - the skin looks mostly smooth - and they are covered in calluses and scratches. (I assume the scratches are unique, but probably not a permanent identifying pattern!)
Personally, my fingerprints often fail to read the same way over some weeks of dry climate, especially if I've done any manual labor. And before you say anything about dry skin, putting lotion on is a temporary DoS attack on the fingerprint scanners I own, so that's not great either.
I can re-train the fingerprints, and they work for some weeks again.
A more successful workaround I've found is teaching the machine the side of my finger, not the pad.
Thanks for the tip. I have iPhones and iPads that have fingerprint readers and they all invariably fail to read my prints after a while (And I have 3-4 prints stored per finger). It’s frustrating. Sometimes it’s due to dryness but other times I have no explanation for.
Surely people are born without hands — can they not get passports? A driver's licence I would perhaps understand, but certainly they should be able to be issued a passport?
Prior to the pandemic my phone could be reliably unlocked with my fingerprint, and I could access my bank accounts and confirm transactions with them.
Since the pandemic started, the extra regular and thorough hand-washing caused my phone to stop recognising my fingerprint most of the time.
Now to unlock I usually have to use another method instead, and some financial apps won't let me in or let me confirm transactions at all because they require the previously registered fingerprint, so I have to use the web version of those services.
Had to deal with fingerprinting for a couple work gigs. One thing that really messed up the process was me putting in bathroom tile. Between me using my fingers to smooth the grout and other manhandling/construction - I apparently wore off the prints enough to cause issues with the reader.
Same thing happens to me when manually sanding and doing any concrete work without gloves. The sandpaper and concrete acid remove enough of my fingerprint so as to not be recognizable anymore.
Most of these employees work in the fields handling heavy equipment. Most of them had no fingerprint since they'd been worn out. We ended up using palm recognition.
The idea that "all people have a unique fingerprint" is a widely believed falsehood.