3 ms·

What technical countermeasures do you suggest? Check out what this guy has to say about resistance: http://youtu.be/3ftfEXxFC4Q?t=28m53s http://youtu.be/3ftfEXx

by jafaku 13y ago

What technical countermeasures do you suggest? Check out what this guy has to say about resistance: http://youtu.be/3ftfEXxFC4Q?t=28m53s http://youtu.be/3ftfEXxFC4Q?t=28m53s

fiatmoney 13y ago
Frankly, I'm hoping that they can come up with some, and some other people can come up with ways to make such measures socially accepted and encouraged. The examples he gives are specifically the kinds of things that there are social solutions to - eg making it Not Socially Acceptable to wear mobile face recognition systems around in public. If you're going to devolve into a DDR-style police state, there's very little technically that can prevent that from happening. What can happen is that you force an enormous investment of resources to maintain it, instead of giving the ability to a tiny cabal. If you're forcing the recruitment of a demographically significantly portion of the population (somewhere 2.5% to 7% in the case of the Stasi) in order to maintain the security apparatus, there's much more room for debate over whether to build it or dismantle it than if it just requires a few billion dollars and a fistful of NSLs.
visarga 13y ago
I have thought about it. We can't make use any more of Google, Microsoft, Yahoo Twitter and FaceBook. We need a new system, one built with anonymity and lacking a central point of failure. Let's say we invent a protocol and write some software clients for it. It should do anonymous search, browsing, messenging, social networking, video calls and video streaming. It would hide the origin of a request and make it uncertain about who does what. For example, if 1000 people use a common VPN server to access Google, and they all come out as the same IP address, it can become more difficult to know who searched for what. All that is needed then is to browse in Safe Mode without logging in. This kind of ideas and other (distributed hash table, using bittorrent like systems for moving data around) could make a nice private communication tool. We just need to make a clear separation between what can be said on FB and what can be said on Anonymous net.
- coldpie 13y ago
  I've been thinking about this, too (I think something built on top of Freenet is the answer, btw). But the trouble I keep running in to is: no one is going to use it. Relatively few people want to hide what they put into the Internet. Everything on Facebook is effectively public, and look how much information gets placed there. We can invent wonderful and secure ideas all day, and have been since the '40s. But what use are they if no one uses them? We need to break this down like any traditional problem and remove irrationality. What, exactly, is the problem we want to solve? What prevents existing technology from solving that problem? Edit: I should add, I'm somewhat interested in discussing this seriously, if you wish to take it off HN.
  - DanBC 13y ago
    > We need to break this down like any traditional problem and remove irrationality. What, exactly, is the problem we want to solve? What prevents existing technology from solving that problem? 1) We want Ann and Bob to be able to communicate. Both Ann and Bob must know that they are actually talking to each other, and not to Eve pretending to be Ann or Bob. Eve must not be able to understand what Ann or Bob are saying to each other. 2) We want Anonymous_Person to be able to send a document to Bob. We don't want Eve to be able to work out who anonymous_person is. We don't care if Eve can read the document. These two are already possible, but hard. Subtle flaws can mean totally broken. Huge amounts of money and research and effort goes into making computer OSs and interfaces easier to use, but there is very little happening around making cryptography and anonymity easier to use. Thus, good quality, short, readable, documentation would be very useful. And much better interfaces and workflow for existing software would also be helpful. 3) We want Anonymous_Person to be able to send a message to Bob. We want Anonymous_Person and Bob to be able to know they are talking to each other, and not to Eve pretending to be one of them. We want to make sure that Eve cannot read the message. And we want to make sure that Even cannot tell who anonymous person is. This is very hard. It requires advanced knowledge, and strict discipline, and very careful use of complicated tools. Then there are a bunch of expectations. "We cooperate only when given a validly formed legal document; and we resist those if we think they're too broad" is one. Perhaps some kind of Privacy Charter? A bit like the GNKSA[1] (Good Net Keeping Seal of Approval) - and companies can have a checklist of things they do to meet the charter. [1] (http://www.gnksa.org/ http://www.gnksa.org/)
eLobato 13y ago
There's this project in northeastern Spain https://guifi.net/ https://guifi.net/, which basically advocates for a distributed network set up by the citizens. As long as your infrastructure is run by a superior power, you can only hope encryption/tor/ssl, etc.. will work properly. Of course if you are sending information to hosts outside the network that information is prone to be subject to surveillance of some kind.

by jafaku 13y ago

What technical countermeasures do you suggest? Check out what this guy has to say about resistance: http://youtu.be/3ftfEXxFC4Q?t=28m53s http://youtu.be/3ftfEXxFC4Q?t=28m53s

fiatmoney 13y ago

Frankly, I'm hoping that they can come up with some, and some other people can come up with ways to make such measures socially accepted and encouraged. The examples he gives are specifically the kinds of things that there are social solutions to - eg making it Not Socially Acceptable to wear mobile face recognition systems around in public. If you're going to devolve into a DDR-style police state, there's very little technically that can prevent that from happening. What can happen is that you force an enormous investment of resources to maintain it, instead of giving the ability to a tiny cabal. If you're forcing the recruitment of a demographically significantly portion of the population (somewhere 2.5% to 7% in the case of the Stasi) in order to maintain the security apparatus, there's much more room for debate over whether to build it or dismantle it than if it just requires a few billion dollars and a fistful of NSLs.

visarga 13y ago

I have thought about it. We can't make use any more of Google, Microsoft, Yahoo Twitter and FaceBook. We need a new system, one built with anonymity and lacking a central point of failure. Let's say we invent a protocol and write some software clients for it. It should do anonymous search, browsing, messenging, social networking, video calls and video streaming. It would hide the origin of a request and make it uncertain about who does what. For example, if 1000 people use a common VPN server to access Google, and they all come out as the same IP address, it can become more difficult to know who searched for what. All that is needed then is to browse in Safe Mode without logging in. This kind of ideas and other (distributed hash table, using bittorrent like systems for moving data around) could make a nice private communication tool. We just need to make a clear separation between what can be said on FB and what can be said on Anonymous net.

coldpie 13y ago
I've been thinking about this, too (I think something built on top of Freenet is the answer, btw). But the trouble I keep running in to is: no one is going to use it. Relatively few people want to hide what they put into the Internet. Everything on Facebook is effectively public, and look how much information gets placed there. We can invent wonderful and secure ideas all day, and have been since the '40s. But what use are they if no one uses them? We need to break this down like any traditional problem and remove irrationality. What, exactly, is the problem we want to solve? What prevents existing technology from solving that problem? Edit: I should add, I'm somewhat interested in discussing this seriously, if you wish to take it off HN.
- DanBC 13y ago
  > We need to break this down like any traditional problem and remove irrationality. What, exactly, is the problem we want to solve? What prevents existing technology from solving that problem? 1) We want Ann and Bob to be able to communicate. Both Ann and Bob must know that they are actually talking to each other, and not to Eve pretending to be Ann or Bob. Eve must not be able to understand what Ann or Bob are saying to each other. 2) We want Anonymous_Person to be able to send a document to Bob. We don't want Eve to be able to work out who anonymous_person is. We don't care if Eve can read the document. These two are already possible, but hard. Subtle flaws can mean totally broken. Huge amounts of money and research and effort goes into making computer OSs and interfaces easier to use, but there is very little happening around making cryptography and anonymity easier to use. Thus, good quality, short, readable, documentation would be very useful. And much better interfaces and workflow for existing software would also be helpful. 3) We want Anonymous_Person to be able to send a message to Bob. We want Anonymous_Person and Bob to be able to know they are talking to each other, and not to Eve pretending to be one of them. We want to make sure that Eve cannot read the message. And we want to make sure that Even cannot tell who anonymous person is. This is very hard. It requires advanced knowledge, and strict discipline, and very careful use of complicated tools. Then there are a bunch of expectations. "We cooperate only when given a validly formed legal document; and we resist those if we think they're too broad" is one. Perhaps some kind of Privacy Charter? A bit like the GNKSA[1] (Good Net Keeping Seal of Approval) - and companies can have a checklist of things they do to meet the charter. [1] (http://www.gnksa.org/ http://www.gnksa.org/)

eLobato 13y ago

There's this project in northeastern Spain https://guifi.net/ https://guifi.net/, which basically advocates for a distributed network set up by the citizens. As long as your infrastructure is run by a superior power, you can only hope encryption/tor/ssl, etc.. will work properly. Of course if you are sending information to hosts outside the network that information is prone to be subject to surveillance of some kind.