One aspect of biometrics - mostly fingerprints, I'll admit - that I don't see discussed very often is the issue of injury, (Note, this is based entirely on my own personal experience with a particular device several years ago, and may not generalise, but I think the general idea is valid).
Let's say I locked my phone with my thumbprint. But then I hurt my thumb chopping vegetables or mountain-climbing or whatever, and for a few days it's wrapped in a bandage to stop myself bleeding all over the place. But that means that in that time it's that much harder to unlock my phone. Yes, in theory you can use your other thumb, if prints are the same across digits (are they? I'm not sure - I know my phone was twitchy at the best of times and definitely didn't work), but that can be awkward switching hands etc. And what if - and this is quite niche - the injury is sufficient to permanently change your print, due to scarring or something? And that's on top of the demonstrated issues if your hands are dirty or wet or whatever.
You'd need some kind of backup, which is fine, but then you just have another potential weakness since any system is only as secure as its weakest point.
I'm not saying biometrics are useless. It's totally plausible that for most people, most of the time, they're a strong improvement in both security and convenience - I haven't done the research on that topic. But I rarely see that view discussed. Maybe because it's not actually a major issue, I don't know.
Prints are not the same across digits, so you'd have to register your other finger(s) in the scanner, which might be a convenient solution for your phone but may get more complicated with online services or physical spaces like biometrically accessed offices or classes. While there may be similarities in the general structure and characteristics of fingerprints across different digits of the same individual, the specific traits like ridges are different for each finger, even across the same person or identical twins.
Regarding the discussion on permanent injury, I think it's unlikely to be raised on a big scale due to the fact that the privacy community is already a niche group. Moreover, considering the relatively low(ish) overall probability of somebody sustaining a lifetime injury, combined with the niche nature of the community, the statistical likelihood becomes (niche-squared) very low. A normal person would just be focused on finding a workaround and would most likely not have privacy in mind: I don't think a one-eyed individual's first thought - after going through all the trauma - would be to vent on some privacy forum about their inability to access a retina-unlocked toaster.
Nevertheless, it would be interesting to hear any stories about something like this happening, whether to raise awareness or just for curiosity's sake.
Oh, for sure, this is not a major problem. Like I said, for most people, most of the time, it's entirely plausible that biometrics are a big help. Just an aspect I don't see discussed.
Most of the time if these issues do come up, they're temporary and mild. Like me cutting my thumb - in a few days it'll be fine, and if I thought to do both thumbs then I'm almost certainly fine, and my concerns are relatively short-term, and if I did think ahead it's mild. But I would guess that few people think to do that when setting these up (although I'd be curious to see if anyone has examined this).
I agree that the idea of permanent injuries (e.g. scars) are niche, and that the "privacy community" is niche, but given the commonality of biometrics to unlock phones I think this is a bigger concern than we might naively assume. I haven't done a rigorous review, but I found one study that implied that 69% of iPhone users use a fingerprint to unlock their phone (https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4969&context=sis_research p. 8), so I would imagine this is a problem. On the other hand, that's a large enough number that if it was a problem I would imagine it would be more established, so maybe my analysis is wrong.
One aspect of biometrics - mostly fingerprints, I'll admit - that I don't see discussed very often is the issue of injury, (Note, this is based entirely on my own personal experience with a particular device several years ago, and may not generalise, but I think the general idea is valid).
Let's say I locked my phone with my thumbprint. But then I hurt my thumb chopping vegetables or mountain-climbing or whatever, and for a few days it's wrapped in a bandage to stop myself bleeding all over the place. But that means that in that time it's that much harder to unlock my phone. Yes, in theory you can use your other thumb, if prints are the same across digits (are they? I'm not sure - I know my phone was twitchy at the best of times and definitely didn't work), but that can be awkward switching hands etc. And what if - and this is quite niche - the injury is sufficient to permanently change your print, due to scarring or something? And that's on top of the demonstrated issues if your hands are dirty or wet or whatever.
You'd need some kind of backup, which is fine, but then you just have another potential weakness since any system is only as secure as its weakest point.
I'm not saying biometrics are useless. It's totally plausible that for most people, most of the time, they're a strong improvement in both security and convenience - I haven't done the research on that topic. But I rarely see that view discussed. Maybe because it's not actually a major issue, I don't know.
That is an interesting perspective,
Prints are not the same across digits, so you'd have to register your other finger(s) in the scanner, which might be a convenient solution for your phone but may get more complicated with online services or physical spaces like biometrically accessed offices or classes. While there may be similarities in the general structure and characteristics of fingerprints across different digits of the same individual, the specific traits like ridges are different for each finger, even across the same person or identical twins.
Regarding the discussion on permanent injury, I think it's unlikely to be raised on a big scale due to the fact that the privacy community is already a niche group. Moreover, considering the relatively low(ish) overall probability of somebody sustaining a lifetime injury, combined with the niche nature of the community, the statistical likelihood becomes (niche-squared) very low. A normal person would just be focused on finding a workaround and would most likely not have privacy in mind: I don't think a one-eyed individual's first thought - after going through all the trauma - would be to vent on some privacy forum about their inability to access a retina-unlocked toaster.
Nevertheless, it would be interesting to hear any stories about something like this happening, whether to raise awareness or just for curiosity's sake.
Oh, for sure, this is not a major problem. Like I said, for most people, most of the time, it's entirely plausible that biometrics are a big help. Just an aspect I don't see discussed.
Most of the time if these issues do come up, they're temporary and mild. Like me cutting my thumb - in a few days it'll be fine, and if I thought to do both thumbs then I'm almost certainly fine, and my concerns are relatively short-term, and if I did think ahead it's mild. But I would guess that few people think to do that when setting these up (although I'd be curious to see if anyone has examined this).
I agree that the idea of permanent injuries (e.g. scars) are niche, and that the "privacy community" is niche, but given the commonality of biometrics to unlock phones I think this is a bigger concern than we might naively assume. I haven't done a rigorous review, but I found one study that implied that 69% of iPhone users use a fingerprint to unlock their phone (https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4969&context=sis_research p. 8), so I would imagine this is a problem. On the other hand, that's a large enough number that if it was a problem I would imagine it would be more established, so maybe my analysis is wrong.