Comment by nickpsecurity | Hacker News Reader

Comment by nickpsecurity | Hacker News Reader

This is utterly ridiculous, possibly even subversive, given two things:

(a) The number of intentional leaks in Microsoft products (esp new Window's) that might be inspired by NSA surveillance

(b) The leaks showing German BND and Five Eyes cooperate very closely on SIGINT with BND letting them use selectors that no patriot of German government or industry should've tolerated.

Many articles, esp Der Spiegel, also indicate that German intelligence is lap-dogging so hard in an attempt to join Five Eyes club. Additionally, remember that the TAREX teams in ECI leaks were focusing on Germany, South Korea, and China. Yeah, not all the terrorism they talk about on TV: one opponent and two seeming partners. What do all three have in common? They're among the biggest economies competing with American (and Five Eyes') business interests. Makes sense among scheming intelligence agencies to use assets against them to get privileged few tight with Washington a competitive advantage and do counter-intelligence against their agencies doing the same.

So, Microsoft are either a bunch of idiots that have no awareness of Snowden leaks or reporting on the situation with NSA and Germany. Or they knew that while subverting their European offering to give NSA every chance to hit the customers' data as a favor to them to avoid pressure and keep lucrative contracts. I'm betting on the latter to be safe and continuing a boycott of Microsoft tech where possible.

That's because it's really a tactic to avoid EU privacy laws, not to restrict the NSA. See contractual solutions here:

http://www.europe-v-facebook.org/EN/Complaints/PRISM/Respons...

If Microsoft would just keep the data there, that would be a good thing. But I suspect it will transfer the data from Deutsche Telekom back to US servers, where the US government will be able to request it anyway. I'm hoping Microsoft won't do that, though.

Very possible. Could be both.

> But T-Systems will act as a “trustee” of the facilities, with Microsoft insisting its employees will have no access to the data held at the facilities without the German company’s permission.

If Microsoft were actually interested in privacy, they'd design their products in such a way that Microsoft would have no technical ability to get to users' data without the end user's permission. A true zero knowledge data center whereby Microsoft does not have the technical ability to restore users' data is the only viable solution.

It seems to be more of a tactic to get out of the current law case against them: US vs. Microsoft where the US wants access to european servers from Microsoft to get the messages from a european to a european which discuss a drug crime.

Exactly. The only real solution is client-side encryption. If the encryption is implemented correctly, it does not matter where it is hosted.

Client side encryption, and no physical access to the outside... and...

Considering that many encryption systems around seem to get broken, I don't think that's the only solution needed.

You can still learn a lot about a social network just by analyzing traffic patterns and meta data. Encryption by itself is not a panacea.

The real solution is high assurance systems: those designed in a rigorous way with an argument that they'll remain secure against any known classes of attack. That's every layer of the stack and sideways as my proprietary framework [1] used to show. Here's an example [2] of what that's like applied to Tor. Ottela's Tinfoil Chat is only one I know that applies high assurance at the design level to eliminate risk of High Strength Attackers. He took a lot of feedback from us on Schneier's blog and applied it well.

Client-side encryption using C libraries on Linux or whatever is just asking to be hacked. Best will be secure hardware, drivers, kernel, and trusted components. See CHERI processor [4], EROS documentation [5] (esp "Principle-driven...") and Mikro-SINA VPN [6] for examples of The Right Thing in security-critical development. Need more like that.

[1] http://pastebin.com/y3PufJ0V

[2] https://www.schneier.com/blog/archives/2014/09/identifying_d...

[3] https://github.com/maqp/tfc-otp

[4] https://www.cl.cam.ac.uk/research/security/ctsrd/cheri/

[5] http://eros-os.org/eros.html

[6] http://genode-labs.com/publications/mikro-sina-2005.pdf

In other words, the pieces may have changed hands but the great game continues its course...

This is a bid to place Microsoft above reproach. They get to play the "we did everything in our power to protect privacy" card, while still remaining fully compliant with government back-door policies. Now the media runs with this, and eventually the mindless masses are lulled back to sleep while their rights and freedoms continue to be eroded.

There are a lot of tech companies that can argue that they are doing everything within their power to protect privacy from government actors (e.g. Apple, Yahoo, but not, say, Verizon[1]). It just so happens that "everything within their power" might be extremely little, if they are required to maintain their existing structure and comply with the letter of the law.

At some point our options really become: relying on tech providers without presence in "unsafe" jurisdictions (if there are any), relying on tech providers willing to disobey the law (and ideally capable of convincing us they disobey them in our favor. Via: open-source, audits, etc) or pursuing reform of those laws.

[1] https://www.eff.org/who-has-your-back-government-data-reques...

Hi everybody. Just your friendly neighborhood Linux nerd. Use Linux, it's open source.

Which gets you what against the NSA, BND, and mass collection? Nothing by itself. Solving these problems is a tad more difficult than merely using Linux. ;)

> Which gets you what against the NSA, BND, and mass collection?

A step away.

> Nothing by itself.

Not so.

> Solving these problems is a tad more difficult than merely using Linux.

True. But it's swallowing the red pill, just something you have to swallow to know how deep the rabbit hole goes.

"A step away" = still there. You're just being contrarian.

"Not so." Prove it. Linux still shows your communications on plaintext online by default. They have zero days in popular apps and kernel for Linux. They have attacks below firmware if you get clever with OS security. So, no, using Linux by itself doesn't change anything for bulk collection as proven by the Snowden leaks. It can be an improvement over Windows for targeted collection but will still be hit in default state. And most people hearing "Use Linux" aren't qualified to get it anywhere near bulletproof. So, they'll be easy targets.

Using any standard, mainstream system when nation-states are after you equals much higher chance of getting your box owned. It's just economics: they focus on attacks w/ highest ROI. Windows, Mac, and Linux kernels are highest ROI for desktop. People on an Amiga are probably less likely to get hacked by automated system despite inferior security just because of its obscurity and attackers' focus. It's why my old trick of BSD or UNIX on non-popular ISA's (eg Alpha) stopped so many attacks despite their apparent sophistication.

I love Ubuntu and haven't touched Windows in 2+ years, but I don't delude myself into believing that Ubuntu being Debian-based affords me protection against mass surveillance.

I guess it depends on the method of surveillance. If your data is being collected by any 3rd party then you might as well assume that any vague yet menacing government agency can access it since it's just a subpoena away. So all the tracking and analytics that MS/Apple/Google/Damn near every tech company put on their services is really the problem.

The only method of securing your data is to make sure it never leaves your computer without your consent and using Linux is one of many possible things you can do to ensure that. Sure, if you're running Skype, Chrome, Steam, etc. on your Linux laptop then the benefit is pretty marginal but still technically better than if it was on Windows.

- http://www.gnu.org/philosophy/open-source-misses-the-point.h...

- http://www.gnu.org/gnu/linux-and-gnu.html

GNU and links like those consistently miss the point when the discussion is about security. OSS or FOSS don't matter as much as review by qualified, trusted party with trustworthy distribution. As explained here:

https://www.schneier.com/blog/archives/2014/05/friday_squid_...

Might explained why a few closed source products achieved high assurance security (B3/A1/EAL6/EAL7) that survived years of NSA pentesting while FOSS... see Snowden leaks. You can get those freedoms with proprietary models via contracts, esp with non-profits. Most just are about money though. Likewise, FOSS communities rarely deliver high-quality, maintained software and never deliver highly assured software.

So all this philosophical crap serves nobody if we're talking about making software trustworthy. For that, we need capable people building it while utilizing every security-enhancing method we know, trustworthy people reviewing it whether closed or open, ability to know you possess what was reviewed in that configuration, and continual maintenance esp of bugfixes. That's the baseline we need to push whether FOSS or a proprietary version of open-source.

Reply on news.ycombinator.com