I don't understand this line of thinking regarding the big encryption debate. Yeah it sounds good if you don't know anything about encryption. "Look, he's being reasonable. Why is everyone saying he's wrong instead of striking a balance".
The problem is it's a binary problem. There is zero way to strike any type of balance here. You either encrypt communication to prevent others from accessing it or you provide a backdoor or escrow key to access the data in which case all encrypted communications are now insecure.
There isn't a middle ground there. If there is I have yet to see anyone suggest it and I can't think of a technical way to do it. Too many attack vectors when you purposefully impose multiple party keys or backdoors.
I think some people in the tech world are 4th-amendment absolutists. That's not a new thing; in Cryptonomnicon there was the electromagnetic doorframe so strong that it would wipe drives passing through it. I don't know that it's a position I hold myself, but it's one I definitely respect.
But we have a multi-century history of the government being able to read people's documents once they have a warrant. A lot of crime-fighting makes use of this. Whatever the intricacies of the technology, there's a status quo that I think we should at least acknowledge.
So I think the rhetorical balance we need to strike as an industry is along the lines of, "Yes, we totally get why you would want to get into that terrorist's phone. We wish we could find a way to give you access to just that without compromising everything. But it's not like a wall we can put a special door in. It's like a balloon. The moment it loses integrity, it's worthless. We want to help you, but we can't."
When the message instead comes across as absolutist, I think we're in danger of losing public support. Without that, we'll have a very hard time resisting government demands for access.
I'm not sure trying to increase the complexity of the issue just to make it sound like the technology community doesn't see this as binary is a good solution. I think what we need to do is education people around how encryption works.
If we're wishy-washy about the whole thing it may come off as something we can possibly do in the end. I think education is the only way. But that's just my opinion.
Some people on this are absolutists, and some aren't. If we could create a safe back door, some would be in favor of doing that, and some wouldn't. That's distinct from us not having a way to create a safe back door.
One problem I see us having (and I've seen this on Facebook) is that some people think that we are saying we can't create a safe back door because we don't want to. And I think they're getting that notion because of the people in the absolutist camp.
I've seen a tiny handful of comments like that and other comments about how they're fine if it isn't completely safe in order to "protect the country". Is there any data to even know for sure what the majority of common people think about encryption and backdoors?
I'm not convinced a near majority simply thinks that SV just "doesn't want to". Well except in maybe the current Apple vs FBI case as the FBI and some of the media are kinda shouting that. Perhaps that'll sway the tides and make more people think in that way, however?
Either way I still think trying to educate the populous is the best way to undermine the entire idea of escrow keys / backdoors.
When people can't engage on the substance, they fall back on human heuristics. E.g., bedside manner is important in doctors because people will generally stop listening to doctors they don't find trustworthy.
You're welcome to try to educate the populace on the subtleties of crypto. My guess is that the number of people who will sit still for that is pretty small, but maybe I'm wrong. In case I'm right, though, I think a useful backup would be displaying empathy for the (reasonable) societal goals that are driving this push for back doors.
There is a middle ground and where it lies is directly related to the assessed risk. To support a mechanism is a binary choice but its design is not.
In the San Bernardino example, an unexpected potential adversary (Apple) has emerged. An unexpected vector is being discussed which has nothing to do with the encryption mechanism itself. Perfect cryptography doesn't imply a perfect cryptosystem and it sure doesn't imply perfect privacy.
The balance that needs to be found is around the assistance which is rendered to warranted agencies to enable their access to data. IMO vendor supported key material attacks are not an unreasonable solution but obviously that mechanism can go horribly wrong if the software is leaked and it is not adequately designed. The same can be said for signed updates in general though. This is a hard technical problem but it's not a binary choice where the agency has some magic key they can use wherever they want or they have nothing. If they can be enabled on a case by case basis such that it is cost effective for them to do the rest, that might be enough.
Obviously this is all based on the assumption of a robust local judicial process. However, whatever decision making process occurs needs to take into account global implications. Other commenters here and elsewhere have mentioned what this could mean for agencies in other countries where judicial process may be less robust. Many have commented about a lack of transparency in our own. This is a separate but very closely related concern. There are also considerations for the open source community and how precedent like this might be applied to them. It's a hard problem but it's not binary.
It's also a binary condition whether a company should produce an unsafe fork of their own software and make it available to any government. If they should, then in effect all governments will use it against their adversaries, foreign and domestic.
Why is it not unreasonable? You're compelling companies to write custom software explicitly for law enforcement use. Let's say you own a company that creates product X. Now the FBI is compelling you, through the court system, to develop a completely custom version of product X to let them access it or use it in some way. Sure they're paying you but you now have to redirect resources to handle this and you likely hired people to do things other than work on a government project. This ultimately can hurt your company and its market position simply by consuming your resources.
This is unprecedented outside of war times. It's hard for me to imagine anyone being okay with compelling a company to become a government contractor.
> It's a hard problem but it's not binary.
It absolutely is. The "middle ground" you pointed to isn't a middle ground; it's unconstitutional. I have yet to hear a technically viable AND constitutional "middle ground" in this debate.
Did you know encryption technology is not limited to Apple? It's not even limited to big companies. Anyone with a book or internet connection can implement encryption.
This isn't plutonium, which is hard to come by. Any law that puts a backdoor on phone's encryption schemes will fail to stop criminals from hiding their communications. They'll simply use another device or download a different app that does give them encrypted communications.
You and Obama are both correct to point out we should be seeking a balance to maintain our security. However, your calculation of that balance is missing some key elements, such as the one I mention above, which is that this is a gigantic game of whack-a-mole. It's the biggest one you could ever play. There's no way you can censor the internet or ban certain kinds of software. It's the same as censoring speech. Encryption is just code. The policy Obama is implicitly proposing simply won't work.
I can see you've thought about this so I'll give you one example where I think we should unlock a phone.
In the movie-like scenario where there's a nuclear weapon hidden somewhere, and an encrypted phone holds its location, then I would expect the NSA and FBI and every computer in the world to sick their processors on copies of that phone's data in an attempt to break the encryption. I believe this is already described as something the NSA has in place with a program called Bullrun. It seems likely to me that the FBI is unable to make use of this technology because decryption is so laborious, even for the mass of computers that the NSA has built up, and even for all of Google's machines. They just don't have enough to decrypt every criminal's phone. But I believe they could decrypt the movie-like scenario nuclear-weapons-hiding phone.
So we're covered in the most extreme case. But in reality there are all kinds of ways to discover someone's password, and you don't always need a huge server farm to get at it. For example, software has weaknesses that are only known to the NSA. No software is 100% impregnable. Even Snowden admits this.
We've existed as a society for a long time without encryption. Bad actors who collaborate exclusively online using encrypted technology, and who never meet in person, are few and far between.
Unless the US government has some major secrets to reveal about the balancing factors of maintaining public security, then on balance, trying to force backdoors into phones will make us all less secure. This forced weakness will be exploitable by every hacker in the world, at no cost, and from positions which are not governable. And the phone manufacturer will be required to keep that weakness as-is because to fix it would be to break the law.
Higher priority freedoms:
1. Freedom of religious choice and freedom from religious persecution. Freedom of religious assembly.
2. Freedom to immigrate to the US.
Lower Priority Freedoms:
1. Freedom to purchase guns and ammunition
2. Freedom to use encryption to keep personal communication private
The government isn't saying that no one should be able to encrypt data. They just thing that, in some cases, it should be done in such a way that certain data can be read by use of a centralized master key.
This, of course, presents certain problems. What if that master key leaks?
But it's not binary. Data secured this way is absolutely safer than not encrypting something at all. "Is it safe enough?" is certainly a question that people can disagree on.
It's not binary.
So if you're doing US national security, you cannot accept the Obama position; any technology to which our "good guy" law enforcement could get access is guaranteed to also be a technology to which a "bad guy" foreign enemy combatant could similarly gain access.
If you're a major US corporation, you cannot accept the Obama position; any technology to which our "good guy" law enforcement could get access is guaranteed to also be a technology to which a "bad guy" foreign intelligence/espionage agent could similarly gain access.
So as far as that is concerned, it is binary. You can either be secure against foreign threats, in which case you are also as a side effect impenetrable to the local "good guys". Or you can be penetrable by the local "good guys" and also as a side effect penetrable by the foreign "bad guys".
The argument advanced against Obama's position is that A) it fundamentally fails to acknowledge this reality, and B) seems to argue for deliberately compromising national security in order to... preserve national security? In other words, it is not only unrealistic, it is in fact nonsensical and self-contradictory.
Again, security & encryption aren't binary. It isn't on or off. Different levels of protection are appropriate for different levels of threats. The lock on my mom's suburban house is a lot weaker than the lock on the vault holding the gold at Fort Knox.
Similarly the kind of security appropriate for consumer grade cell phones might be different than the kind appropriate for guarding US national security secrets.
You have a very narrow view of what information is sensitive. Do you know how many Americans work either directly in, or in industries associated with, defense? How many of them use Apple products as part of their work?
Or let's take a more down-to-earth example: I work for a company in the health-care space. We have health records on thousands of people, and as we expand we'll have more health records on more people. With your approach, where should we draw the line? Would it be OK to use security bad enough that, say, an Eastern European organized-crime ring could break in? Or do we need to defend against them and not against the KGB? How do we figure out what level of tools to use to do that? How do we figure out whether what we have now will still be mafia-proof in five years?
Oh, and keep in mind that legally we have to protect that data, and improper access to it can not just shut down the company but send me and my co-workers to jail. How much would you be willing to compromise with that on the line? If the potential consequences to you were jail time, how much on the side of Obama and the FBI would you be? How much risk of that would you take on in order to make law enforcement's life a little bit easier?
And in case you think this is a red herring, remember the San Bernardino shooter... worked for a public-health department.
You should fix that.
The fact that it's an iPhone in the San Bernardino case is kind of a red herring; the precedent it would set is not "iPhones and only iPhones and not any other type of computing device are subject to law enforcement unlocking". The precedent it would set is that any encrypted system must be built to be decrypted on demand by law enforcement.
And that gets back to my original comment: a system the "good guys" can decrypt is one the "bad guys" will be able to decrypt too.
Like, I get you. By allowing X which might actually be OK, we might allow X+1 to happen and that wouldn't be OK. So we must oppose X as a practical matter.
I get the practicality of that.
But when you start opposing X as a matter of principle rather than simply as a matter of practicality that's when you've gone off the rails.
Comments like this are exactly why he used it.
And quite frankly, yes, I do hold the security and privacy of data as more important than the terrorism song and dance the government is playing. And I do hold that view absolutely.
The government always leaks keys. Sometimes literally.
http://www.wired.com/2015/09/lockpickers-3-d-print-tsa-lugga...
http://nypost.com/2015/09/20/the-8-key-that-can-open-new-yor...
Plus, any key that the government can use without your permission, they can illegally abuse. And if there's one thing we've seen proven recently, it's that such abuse will occur.
That's still pretty safe.
* Minors cannot buy guns.
* Convicted felons cannot buy guns in most circumstances.
* There are numerous locations in which it is illegal to bring a gun.
* There are numerous types of guns that you are not allowed to own.
It's not binary.It's still not broken in any technical sense, but if I don't want a third party to read my data and they are still able to do so, it's a stretch to call it encryption.
But Google itself can access and read your email. It does so in an automated fashion to target ads at you.
So is your gmail encrypted? Well...kind of.
It's not binary.
With a government-held master key, there is no such contract.
And besides, the existence of a contract or not isn't really my point. My point was that encryption isn't an all or nothing situation. You don't either have completely ironclad security where you and only you can access your data or nothing at all.
There are degrees of technical protection. That was my point. The black and white thinking about this issue is wrong.
Isn't the point of discussing this that you actually don't have anything that defines that contract? In terms of the fourth amendment, it seems like there is quite the opposite contract in place already. Also, as an Apple product user outside the U.S., this supposed contract is not something that I ever agreed to.
> And besides, the existence of a contract or not isn't really my point. My point was that encryption isn't an all or nothing situation. You don't either have completely ironclad security where you and only you can access your data or nothing at all.
It is an all or nothing situation. That there is only one unauthorized party able to intercept my data using a back door doesn't change the fact that there is an unauthorized party able to intercept my data using a back door. Whether that is practical for you or not, again, depends on whether you consider the third back-dooring third party authorized.
> There are degrees of technical protection.
I think that maybe the discussion here comes from the fact that you are indeed talking about "technical protection" in general and not encryption specifically. The point of encryption is that only eligible parties can decode the data. If others can decode the data it's not really encrypted, as a matter of definition.
For example, using Google as a mail provider may prove to be poor "technical protection", but the data is encrypted and the authorized parties are part of the terms of service. That the U.S. government can decode my encrypted data without permission may prove to be practical "technical protection", but in terms of encryption it really is a boolean situation.
I lied, there is an award for effort. Wherever you have an "reasonable expectation of privacy", the government must get a warrant or your permission to exercise its search and seizure authority. That's the protection that you're entitled to by our laws.
Even the EFF is acknowledging the weakness of your argument when they try to frame it as an issue about Apple's free speech instead.
"And we agree on that, because we recognize that just like all of our other rights ... that there are going to be some constraints we impose so we are safe, secure and can live in a civilized society."
I'm not suggesting that this means POTUS and the government have the right answers at the moment. Despite that, we can't ignore the important role law enforcement plays in society, the requirements in support of their role, and the complexities surrounding the right to privacy. We need people advocating for the right balance, not just getting each other frustrated.
OP may have worries other than US law enforcement being from another country. This is one of the /many/ complexities in this space.