Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A Back Door to Encryption Won't Stop Terrorists

bloombergview.com

Post reply on HN
Print view View original

231–240 of 282 posts

Re: A Back Door to Encryption Won't Stop Terrorists

#231

Post by Steve44 »

bobdole1234 wrote: post #171
jacquesm wrote: post #101

So, here's my take on all this 'surveillance is good for you'. It more or less proves (to me at least) that the government(s) and the various secret services have absolutely no idea who to monitor specifically. So instead of targeting their operations they want to monitor all of us, just in case something of interest pops out that then allows them to focus their attention. It's a pretty scary thought: just imagine, a…

No, they want to collect everything to find something to hang you with after the fact. There is way too much volume and noise to find threats in the hurricane of hay flying past to find the needles in real time. But if you have someone specific to start targeting, you can quite easily take their whole life apart and find their friends and make all these connections after the fact. The goal isn't to stop attacks, it's…

It helps them to find people who they were in contact with and so get deeper into their network. If they have a log of all the telephone calls they are then able to run a simple "SELECT * from BigTable WHERE 'badphone' = NumberCalled" and then take the investigation from there. I know that's simplified but gives a highly filtered starting point of people of interest.

Another reason not to publicise their successes is that may give an indication of how they are tracking and if the terrorists know how they are being tracked they can change how they operate. For example perhaps they always call directory enquiries to get a number/address and if that is the way in it's in the interest of the spies to not let them know that so they can continue to use that chink.

Re: A Back Door to Encryption Won't Stop Terrorists

#232

Post by hokkos »

TeMPOraL wrote: post #182
afarrell wrote: post #174

Earlier quoted context omitted.

How do you convince someone to refuse to be terrorized though? If I know someone who feels genuine fear when they hear about these sorts of attacks, what do I say to them? "You're an idiot", "here is some math", and "you're giving them power over you" are not effective approaches. How do I convince someone they should try to not feel fear if they strongly believe their fear is valid and that attempts to invalidate th…

I think "here is some math" is a good way, but you need to be really delicate about it and take it slowly. What you're fighting is an emotional reaction that's hard-wired into our brains. The problem stems from them having an invalid view of the world, i.e. perceiving terror attacks as vastly more probable than they really are. Therefore you need to fix their worldview - or rather, help them fix it themselves. Showin…

The problem with those "maths" is they are no longer in favor of not being "terrorized" in some places.

Re: A Back Door to Encryption Won't Stop Terrorists

#233

Post by TeMPOraL »

hokkos wrote: post #232
TeMPOraL wrote: post #182

Earlier quoted context omitted.

I think "here is some math" is a good way, but you need to be really delicate about it and take it slowly. What you're fighting is an emotional reaction that's hard-wired into our brains. The problem stems from them having an invalid view of the world, i.e. perceiving terror attacks as vastly more probable than they really are. Therefore you need to fix their worldview - or rather, help them fix it themselves. Showin…

The problem with those "maths" is they are no longer in favor of not being "terrorized" in some places.

The math is still in favour in the United States and in the European Union. Unless the friend lives somewhere in the Middle East, I don't see a problem with that argument.

Re: A Back Door to Encryption Won't Stop Terrorists

#234

Post by arpa »

fucking_tragedy wrote: post #162
helicon wrote: post #25

The IRA were known to recruit top stem students from universities in Ireland during their campaign to make bombs. Surely an entity as large and as well financed and ISIS would have little trouble finding bright young engineers & technologists sympathetic to their cause to simply build their own encrypted services? And then so much for the spooks 'backdoors'

Having traffic encrypted with illegal algorithms pass through the Internet would make you a target.

How would one determine whether a legal algorithm was used or an illegal one? Even without steganography?

Re: A Back Door to Encryption Won't Stop Terrorists

#235

Post by mike_hearn »

wlesieutre wrote: post #89
mike_hearn wrote: post #70

Earlier quoted context omitted.

Gah. Look, I'm a big fan of end to end encryption and I'd like to see more of it out there, but this kind of post just shows we're collectively missing the point these politician/FBI types are making. There is a huge difference between https://gmail.com type encryption and true end to end encryption. Pretending there isn't will get us nowhere. The first type is sufficient to keep out ordinary criminals, competitors,…

My biggest concern isn't so much that the government could get a warrant with probable cause and intercept a few emails or monitor my web browsing, it's that for all I know (seems pretty likely?) a significant chunk of everything I do online is intercepted and stored forever , encrypted or no. If we allow a ban on secure end-to-end encryption, are we setting ourselves up for a future government to go back and demand…

Just use companies that delete stuff when you ask them to. There's no need for encryption to solve the lifelong-record problem.

Re: A Back Door to Encryption Won't Stop Terrorists

#236

Post by Khaine »

I think one thing we as a tech community overlook is the expectations on the intelligence community. The broader community expects the intelligence agencies to stop ALL terrorists attacks, and thats just not feasible. This drives the intelligence agencies to do more, which is why I think there is a big push for broad dragnet activity.

Re: A Back Door to Encryption Won't Stop Terrorists

#237

Post by Khaine »

pakled_engineer wrote: post #199
criddell wrote: post #167

Earlier quoted context omitted.

> If surveillance works, then Paris, Beirut, Bagdad et al wouldn't have happed. I don't think surveillance is very good at preventing terrorism, but I think it could be useful to law enforcement after the fact to round up co-conspirators. I bet that once the identity of the Paris attackers was learned, the NSA pulled up a graph several levels deep of everybody they communicated it or associated with. It could very we…

They found one of the terrorists phones dumped in a garbage can outside the theatre they attacked, and got the address of their hideout from the messages, so used just regular police work. All these guys were already on watch lists for fighting in Syria and returning just nobody seems to ever watch the watch list.

France has something like 11,500 on the extremist watchlist. I'm not sure its practical to watch and monitor that many people with the staffing levels of French Intelligence without acting like the stasi

Re: A Back Door to Encryption Won't Stop Terrorists

#238

Post by mike_hearn »

Canada wrote: post #206
pakled_engineer wrote: post #199

Earlier quoted context omitted.

They found one of the terrorists phones dumped in a garbage can outside the theatre they attacked, and got the address of their hideout from the messages, so used just regular police work. All these guys were already on watch lists for fighting in Syria and returning just nobody seems to ever watch the watch list.

If that phone was properly secured then those messages wouldn't have been accessible.

True, I thought the same.

But then, if they'd used a CB radio, there'd have been no records on the device at all. You can go both lower and higher tech to avoid leaving a trail.

The fundamental issue our society faces is that the world people thought existed, pre-Snowden, wasn't all that bad. There are controls, a system of laws and warrants, judges have to approve requests for data, and companies hand it out only in cases where it's really justified. Terrorist used WhatsApp to send a message to his buddy before blowing up a football stadium? It is - literally - warranted.

Post Snowden we know that world didn't really exist, it was more a sort of theatre governments put on when they didn't care much about the outcome. So now the data is disappearing so nobody has it. It's not obvious that this is an improvement over restoring the world that was previously thought to exist, but that was apparently too unstable to survive.

Re: A Back Door to Encryption Won't Stop Terrorists

#239

Post by nitrogen »

rnovak wrote: post #165
awakeasleep wrote: post #153

Earlier quoted context omitted.

Sort of the invisible unicorn argument. I can't really say how many invisible unicorns are in the room with me right now. Could be dozens, if they're small, quiet, and quick to get out of the way.

Not entirely, it's more of the "that word 'proof'/'proves' doesn't mean what you think it means" argument. My point is entirely aimed at the linguistic use of "proves" than anything else.

As always, the burden of proof lies with the party making the existence or effectiveness claim. It is on the surveillance proponents to prove their effectiveness.

Re: A Back Door to Encryption Won't Stop Terrorists

#240

Post by nitrogen »

downandout wrote: post #229
jacquesm wrote: post #228

Earlier quoted context omitted.

I don't see this scheme working. It would require a whole bunch of sensors and a chunk of local computing power to make a 'kill' decision (or at least, an attempt to immobilize) on an individual in a crowded space based on the reliable detection of bullets fired from a gun. The kill device could be disabled prior to discharging a fire-arm, it could trigger falsely, it could be made to trigger falsely on purpose and i…

It wouldn't require "a bunch of sensors". Anyway, generally you're referring to technical challenges, all of which could be readily solved.

How would you detect a moving bullet unambiguously?
Post reply on HN