search

dbosk / crocus

Securely and privately verifiable protests
Other
0 stars 0 forks source link

Broadcasting proofs for witnessing #28

Closed dbosk closed 6 years ago

dbosk commented 6 years ago

Can we use ad-hoc networks or bluetooth? The latest spec of bluetooth seems to have increased the limit of 10m to 100m. Can it be used for broadcast?

We can always fall back to publishing them on the Internet? Although it would be undesirable to require an Internet connection due to government DoS attacks.

dbosk commented 6 years ago

There are at least two examples of practical use of ad-hoc/mesh networks

How much should we do? Just point to these saying it's possible, or look into how we must adapt them?

dbosk commented 6 years ago

@sbuc : Can you write a few words about MANETs (file MANETs.tex)? More specifically, why we can assume there is a way for the protesters to communicate locally at the protest without relying on the Internet. Also mention potential privacy problems at this layer, e.g.\ MAC-address tracking etc. which is outside of our protocol (this should cover part of issue #58).

sbuc commented 6 years ago

done, except the MAC tracking. Should be fixed by randomly changing the MAC but need to check how easy that is on phones. For BT broadcast and communication without pairing, this seems probably doable but not easy https://electronics.stackexchange.com/questions/47273/is-bluetooth-communication-possible-without-pairing , but for now I think it's enough to include Briar and FireChat as they do seem to work and it doesn't say anything about pairing. Even if they do pairing somehow it seemed to work in the protest scenario.

dbosk commented 6 years ago

On Mon 26 Feb 2018 21:06:53 GMT, sbuc wrote:

done, except the MAC tracking. Should be fixed by randomly changing the MAC but need to check how easy that is on phones. For BT broadcast

It's currently used on phones, IIRC. However, I recently stumbled on this paper:

[Why MAC Address Randomization is not Enough: An Analysis of Wi-Fi  Network Discovery Mechanisms](https://dl.acm.org/citation.cfm?id=2897883)

and communication without pairing, this seems probably doable but not easy https://electronics.stackexchange.com/questions/47273/is-bluetooth-communication-possible-without-pairing , but for now I think it's enough to include Briar and FireChat as they do seem to work and it doesn't say anything about pairing. Even if they do pairing somehow it seemed to work in the protest scenario.

Yes, I think it's sufficient to refer to Briar and FireChat, although I'm not entirely sure how secure they actually are from tracking. My impression is that they're "made by programmers" and not "made by privacy researchers". In either case, I think that the "network layer" is out of scope for us and that we can assume it can be solved independently.

dbosk commented 6 years ago

I think we can consider this done.