Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

First, it doesn't matter what process governments (plural) have in place to protect backdoor access mechanisms or escrowed keys. That those mechanisms exist at all means they can and will be abused by people who do not go through those mechanisms.

Beyond that:

> If government is monitoring your communications, then there's a pretty dammed good reason, as determined by a judge.

Not even close to true. We have documented evidence at this point that the government is monitoring all communications. They have no discretion.

For the moment, I trust that the government has not yet descended so far that it will arbitrarily arrest or detain everyone who attempts to maintain the security of their information. That doesn't mean I trust the government to unilaterally decide whose information it can access.

Government brokenness is not binary. There's a spectrum from "theoretical perfection" to "absolute tyranny". The US and UK governments are currently attempting to take a step in the wrong direction; that doesn't immediately imply that anyone opposing that step "ceases to be able to function in a society". Or are you seriously attempting to suggest that the government is entirely infallible and incorruptible?

The safeguard for privacy isn't just the court system. That provides a modest amount of protection against abuse of information that was left insecure to begin with, such as information left unencrypted with a third party. However, for information that truly needs security, the privacy of that information is maintained by the system holding it; that system neither has nor needs any concept of "a user other than the owner who somehow has access".

Let's ignore the Apple case for a moment, since Apple fundamentally does have access to the device. Even additions like the more recent security chip have potential ways to bypass security, albeit by dissecting a tamper-resistant chip; that's a tradeoff for supporting insufficiently long user PINs, and wouldn't be needed with a secure passphrase. Whether Apple wins or loses, the much more interesting question is whether challenges arise to the construction of devices that genuinely cannot be accessed by anyone other than the device owner.

As a simple example, consider an encrypted device, using a key and passphrase, where the passphrase uses an appropriate strengthening technique to make it computationally intensive to test a single possibility. Easy to get in if you know the passphrase, but if done right, computationally infeasible to brute force even with NSA-level computing resources and centuries of time.

Laptops regularly use that level of security today (modulo the user's chosen passphrase). Phones should as well.

If I had a perfect eidetic memory, I could memorize huge volumes of information and keep no records of them; in that case, that information is completely secure (against anything except rubber-hose). Encryption provides an amplification mechanism there: instead of memorizing gigabytes of data, I can instead memorize a passphrase, and gain access to arbitrarily large amounts of information that nobody else can access (modulo security bugs, or physical device compromise followed by my continued access).



That those mechanisms exist at all means they can and will be abused by people who do not go through those mechanisms.

Isn't this true about traditional wiretaps as well? Yeah, a corrupt agent could illegally wiretap anyone's phone, but that certainly won't hold up in court.


Yes, that's true. And anyone who cares about the security of their phonecalls should be using encryption rather than POTS; for example, you could use Signal for encrypted phonecalls and SMSes. Warrant or not, unless there's a bug in the implementation, the only way you can tap that communication would be to compromise the endpoints (either electronically, or via a physical bug/tap on the device).


They're monitoring meta-data, not private communications.

And for private comms, they have filters that filter them out for US citizens, as the Snowden leaks show.

LOL when people find out that the Snowden leaks actually show the NSA protecting privacy rights of US citizens, in a top-secret program.


Metadata has critical value. And private communications are being monitored, recorded, and searched as well. (Many people, myself included, also reject the notion that unauthorized collection and recording is fine as long as the subsequent searches and filters are authorized.)

You seem to be taking government descriptions at face value, despite documented evidence to the contrary.

Consider the possibility that the government you seem to place absolute trust in is not, in fact, perfect.


Well I certainly don't trust you and your private army's oversight of me, and I'm sure you don't trust me and my army's oversight of you.

So we make an agreement and build a third-party that has oversight over both of us.

Or do you expect me to trust you?


You seem to be attempting to paint everyone who disagrees with you as someone who stockpiles baked beans and ammo in a bunker. You also seem to have no concept of any middle ground between "absolute trust in the infallibility of government" and "radical anarchist".

Let's make this concrete, instead. Anyone can, today, using off-the-shelf software, store information such that only someone with a passphrase can access it, and such that all the computing power in the world would take centuries to access it without the passphrase.

In such a scenario, I don't expect anyone to "trust" anyone, or to need to. (Modulo the issue of trusting the software to be properly implemented, which is a separate discussion but has some plausible solutions.) Instead, I expect that the software in question will prevent unauthorized access, as it was designed to do.


I, for example, am not a US citizen ;)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: