Do you want skullprint security?

[Unorthodox]

http://www.theverge.com/2016/4/25/11501704/skullconduct-biometric-password-authentication

Conventional passwords are marching to their death, and biometric authentication might be the future, especially with the possibility of widespread fingerprint scanning or facial-recognition technology. And now, researchers in Germany have a new way to give users access to their devices. Their technique, which they call "SkullConduct," relies on the unique way sound passes through each person's skull. Every skull modifies sound differently, and the researchers measured those minute differences to use as an authentication system.

The researchers took 10 participants and put them in a silent room. They recorded skull vibration samples by having them each wear a modified Google Glass that played white noise. The Google Glass application records sound as it plays and converts the data it gathers into a byte file. Each participants's skull vibration sample was recorded 10 times, which the researchers then used to test whether the Google Glass could correctly identify them. The system correctly recognized users 97 percent of the time.

The system correctly identified users 97 percent of the time

The researchers stipulate that their test didn't take into account background noise, like a subway's roar or people chatting nearby. They also say weight gain or hair growth could alter results. Though they're not at 100 percent just yet, it could be cool if our skulls were used for biometric authentication. That is, if people ever legitimately start wearing Google Glass and eyewear computers.

Edit:  Note to self, editing the poll options and hitting 'enter' autoposts...

Thanks to my bizarre news feed...

[Valka]

My skull is none of Google's damn business.

Only my cats get to mess with it.

[Buster's Uncle]

97% isn't even close to good enough - my relatives are the most likely to have the same bones, and that's who I want privacy from most.

[ColdWizard]

They also say weight gain or hair growth could alter results.

Presumably weight loss or a haircut could also alter results? So don't diet or gain weight, don't grow your hair out or shave it off, or you'll potentially lose access to your device? Not mention the whole background noise issue. 97% in a silent room doesn't seem all that great.

[Unorthodox]

Curious, how accurate are other biometric methods? 

Anecdotally, I'd say my iphone's scanner doesn't recognize my fingerprint around the 90-95% of the time. 

[Valka]

They also say weight gain or hair growth could alter results.

Presumably weight loss or a haircut could also alter results? So don't diet or gain weight, don't grow your hair out or shave it off, or you'll potentially lose access to your device? Not mention the whole background noise issue. 97% in a silent room doesn't seem all that great.

And don't get dental work, experience changes in hearing or vision, or sneeze while logging in.

[ColdWizard]

They also say weight gain or hair growth could alter results.

Presumably weight loss or a haircut could also alter results? So don't diet or gain weight, don't grow your hair out or shave it off, or you'll potentially lose access to your device? Not mention the whole background noise issue. 97% in a silent room doesn't seem all that great.

And don't get dental work, experience changes in hearing or vision, or sneeze while logging in.

Didn't think of vision, cursory research indicates that if you have glasses, you need google frames to attach google glass. So even more extra stuff to buy to access devices.

Curious, how accurate are other biometric methods? 

Anecdotally, I'd say my iphone's scanner doesn't recognize my fingerprint around the 90-95% of the time. 

I'd say my Samsung's recognizes my fingerprint about 95% of the time. But it's easily reswipe-able in all conditions. I'm practically never in a silent room, which would assumably be optimal conditions for the skull-sonar.

[DrazharLn]

It's one of those things where false positive and false negative rates are weighted differently, so just saying 97% accurate isn't really the information you need to assess it.

You want a confusion matrix instead.

*science reporting*

[Eadee]

Since no biometric identification method is 100% safe (injuries or illness might make identification permanently impossible)  you'll always have some kind of backup-method like a password. So why not just stay with the password and save the effort for the average user?

Sure, companies or governments, who store valuable secret data or that can give potential dangerous orders through a computer system, might have use for this, but the average person doesn't need that kind of protection for their everyday computer use.

People with high income might think about increased security for their online banking accounts but for everyone else "not being stupid by trusting anyone your bankaccount information, or writing it down" is the all the security one needs in 90% of the cases.

[DrazharLn]

Phones are turning into payment methods, which might increase your desire for better authentication.

Biometrics are also about convenience. The iPhone fingertip scanner is mostly useful because it lets you securely unlock your device very easily.

Security conscious people, would probably rather just use pass codes or use multi factor authentication - biometrics are generally too easy to fake by determined adversaries.