Hacker News Reader

Comment by suprgeek | Hacker News Reader

Many interesting "nuggets" buried in this report. For example:

...A more sophisticated effort, though, relied on intercepting Google Maps queries made on smartphones, and using them to collect large volumes of location information.

So successful was this effort that one 2008 document noted that "[i]t effectively means that anyone using Google Maps on a smartphone is working in support of a GCHQ system."

At this point it is perhaps not wrong to conclude that the whole internet is bjorked by these agencies. Open to snooping and manipulation at any and every level for any user.

It is time for a reboot, this time with much more focus on security.

View on HN

The sad thing is that if I watched these sort of pronouncements in an episode of "Person of Interest" or some other show I would normally laugh them off like I do the infinite zoom/sharpen capability of security video.

Interestingly, things like the zooming and enhancement of video similar to the TV capabilities often portrayed are also now easily accessible. Read about 'super-resolution' [1,2] techniques for more details.

[1] https://en.wikipedia.org/wiki/Superresolution [2] http://research.microsoft.com/en-us/um/people/cmbishop/downl...

I find the two green slides (Capability - iPhone and Capability - Android) the scariest. Hot mic? Kernel stealth? And self protection? How many networks - knowingly or unknowingly - are delivering these things?

That was in 2008. Imagine what else they've been able to jimmy in 6 years!

I'm still waiting on the reveal that they've stored geolocational data at regular timepoints of every X minutes.

> I'm still waiting on the reveal that they've stored geolocational data at regular timepoints of every X minutes.

Considering Apple did that on your behalf I would be surprised if this was not the case.

http://bits.blogs.nytimes.com/2011/04/20/3g-apple-ios-device...

Apple didn't store users location data. Cell tower positrons were cached on iOS devices, and this was turned into a false controversy because sensationalism sells. Citing this piece as if it asserts that Apple collects user location data in the way the parent poster fears the NSA might do is dishonest.

Apple programmed a feature that cached all location data on a file on the iPhone. GCHQ's linked PowerPoint says "If it's on the phone, we can get it". Ergo, GCHQ and the NSA certainly had access to your cached location data if they wanted it. Maybe Apple intended this to be the case, maybe Apple did not, but the mere fact that the file existed is enough that it most certainly could have been scooped up by spooks.

No. "All the location data" was not cached. Only cell tower locations.

"Maybe apple intended this to be the case" is baseless innuendo and has no place here.

"If it's on the phone, we can get it" has separately been shown only to apply when the agencies have physical access to the device to extract data or implant malware.

You seem to want to spread the idea that spies can access the real-time location of your iPhone remotely - which is what the parent post was fearing, but for which there is no evidence. What is your motive here?

Cell tower data is sufficient to determine phone location via triangulation. Additionally, spies can get real-time data from the carrier directly, as can law enforcement.

The parent comment said nothing about real-time access, but if someone has a remote exploit that gives filesystem access (and if jailbreakers can do it, then the NSA can, too), that location data file would provide a detailed history of the phone's location.

No. Triangulation requires near-simultaneous signal strength readings from multiple towers at the time the position is to be computed.

The cell tower cache you refer to does not contain that kind of data, so the data file des not provide a detailed history of the phones location. This has been shown by the people who investigated the file.

An extremely coarse location, to the resolution of cell towers can be obtained from the file, but as we know, that is available to the phone network anyway.

The parent comment talks about geolocation data at regular X minute intervals. This file does not provide that. Nor does it provide any information that the NSA can't get via the cell network about any phone. Indeed the network likely can provide triangulation.

The link you referenced is nothing but innuendo intended to implicate 'Apple' somehow.

I don't know if 5 billion a day translates to the value of regular or X you have in mind, but mathematically - the answer is yes.

http://www.washingtonpost.com/world/national-security/nsa-tr...

> It is time for a reboot, this time with much more focus on security.

Here's a helping hand for those that don't know where to start: http://prism-break.org/

Maps and other apps on the iPhone weren't using HTTPS in 08 (underpowered device, need to squeeze every last drop from battery). They do now however. It's not just a spy agency issue, anyone could have sniffed the unencrypted traffic.

(underpowered device, need to squeeze every last drop from battery)

Really now? Is that the official reasoning for not using HTTPS?

Well, this is the most common argument before Snowdengate I heard against using HTTPS anywhere, not only on mobile devices.

Back then it was.

I just don't recall anything official regarding that line of thought. Direct PR or otherwise. Is there any examples off the top of your head?

"HTTPS is expensive" has been a widespread (stupid) meme since the invention of HTTPS. How old are you?

Old enough to know a non-answer when I see one.

Again I ask: Any specific examples from companies or organizations that implement HTTP(S) in their products stating device power as reason for non-implementation?

I imagine it to be a horrible miscarriage of trust to not use HTTPS. We made the decision early on that handling any personal data not over HTTPS was massively irresponsible - and this is pre-Snowden.

That said, if they have kernel-level hacks or can intercept and decode HTTPS (or sit and listen on say, any AWS server they want), what does HTTPS really matter against the NSA?

Still, totally irresponsible - battery life is a constant struggle, but not enough to even make us consider changing our API client code.

The problem is that HTTPS is very difficult to audit, we just have to trust that it is being done correctly.

How do you know that the apparently random stream of bits is actually properly encrypted and does not leak private data? It would be better to let the OS add the SSL layer and only let apps talk HTTP. This would give the user much more control.

Reply on news.ycombinator.com