4 ms·
I disagree. Fingerprints are privacy locks, not security locks. The lock in your bathroom isn't meant to secure the bathroom. It's just a way to ensure that pe
by Waterluvian 10y ago
I disagree. Fingerprints are privacy locks, not security locks.
The lock in your bathroom isn't meant to secure the bathroom. It's just a way to ensure that people get the message, "please don't enter".
A fingerprint on a phone is a way of saying the same thing. This phone isn't meant for common use, please don't enter.
- yellowapple 10y agoTo expand on your analogy: a username says pretty much the same thing, except instead of a locked door, you have one of those "Do Not Disturb" signs on your door handle.
- saosebastiao 10y agoI mostly cringe at the argument I quoted because it is often misapplied due to how we have historically misapplied passwords. A username is an identity. Historically due to the difficulty of verifying identities online, we have used passwords as a way to do so. And when all we need to do is verify an identity or control basic access levels (the bathroom lock!), a fingerprint is absolutely good enough. But a password is more authorization than authentication: requiring a password is appropriate when you need a conscious decision, not mere identification. Such as for paying for Pokémon toys. So essentially what I'm saying is that I agree with you.
- Waterluvian 10y agoYes, absolutely. :)
- colah3 10y agoThe "deliberateness" or "conciencousness" of a lock is an interesting dimension. A laser cutter I used to work with had two switches, one of which was a safety switch (like [1]) to prevent accidental activation. The goal here isn't security. It's a design decision to prevent accidents. It's almost a kind of intentional inconvenience. [1] Safety switch with cap prevent accidental switching: http://acuteelectrical.com.au/safety-switch/ http://acuteelectrical.com.au/safety-switch/
- Waterluvian 10y agoI bet legal experts could speak to this very well. At what point is someone breaking in?
- justinkramp 10y agoIdentity <> Authentication <> Authority I deal with this in my industry (telecommunications). Just because you've provided proof of identity (eg your phone number, account number), there are still things you're not allowed to do until you've authenticated your identity--and the system determines your authority, to perform an action. This is accomplished through a password, a PIN, etc. On a phone, it's an interesting shift because with a PIN, we essentially bypassed the need for identity and used only a password; regardless of who you are, you can get in if you have the right key. With the move to identity being sufficient to unlock a device, we're saying that just on the basis of identity, the authority that used to come with a password (sans identity) can be granted. It's a 180 degree turn. I don't see a way on my iPhone 6s to require both Touch ID and PIN; it's one or the other. Very few interactions require both, i.e. after a restart it requires the PIN before Touch ID will work. edit: to clarify the difference
- saosebastiao 10y agoYou're demonstrating the difference between authentication and authorization, not the difference between identity and authentication. Notice you use the word authority, which has the same root word as authorization. Authentication is merely the confirmation of identity...it is not the same thing as authorization.
- cookiecaper 10y agoIn the case of a single-user phone, is there a difference? The phone's owner has authorization to do anything, including spend funds they've previously enrolled into the phone's wallet systems, etc., so it's kind of a moot point for the purposes of this incident. You seem to be suggesting we add extra layers here so that merely authenticating as the device's owner is insufficient authorization to conduct some actions, and re-authenticating as the owner by using something they know (secret token like PIN/password) instead of something they possess (finger) will re-grant authorization, but users find this constant re-auth very annoying. Most would probably prefer device makers to allow them to trust the people whom they sleep around rather than input another authentication method all the time. Personal responsibility has to enter into the equation somewhere. My advice to this parent would be to keep their phone and/or body inaccessible while unconscious.
- colah3 10y agoThat's a really nice distinction. It feels like there's a bit of a spectrum, though. The lock on a houses door won't prevent a determined criminal, but it sends a very strong signal and is inconvenient to break. Perhaps we should thing of there as being multiple dimensions to locks, such as security, signaling, effort to circumvent, and convenience? (Perhaps also conscientiousness in disabling, in relation to @saosebastiao's point.)
- Waterluvian 10y agoDefinitely a spectrum. Bike locks are for security, but mainly just to make your bike less convenient than another one nearby. If someone wants into YOUR phone, they'll get it. If someone wants into A phone, a thumbprint may be plenty.
- TeMPOraL 10y agoBeing widespread is a double-edged sword though. The more people have such an inconvenience-based security, the better they are, but if everyone has bike locks, the thieves simply start carrying bolt cutters with them. We're maybe two or three years before majority of smartphones in use have fingerprint scanners...
- threatofrain 10y agoI have a very contrary intuition from literature on crime policy, where the presence of detectable security did not lower home robbery rates for the region, but rather shifted it to unprotected homes and neighborhoods, whereas increases in non-visible security systems lowered crime rates for the region, but didn't provide benefits for the home in question. I recall a similar article on gun ownership in homes, and it noted that gun ownership did not protect any specific home, but rather it decreased home robbery rates for a region.