search

MrChrisJ / World-Citizenship

Globally orientated citizenship with private passport services using available cryptographic tools
Creative Commons Zero v1.0 Universal
402 stars 72 forks source link

Note: GPS data on phones can easily be spoofed). #11

Open chrisjmccreadie opened 10 years ago

chrisjmccreadie commented 10 years ago

Though I agree that GPS can be spoofed and there are new technologies coming out (very early) to deal with this.

I think if you take a photo on your phone and embed the GPS coordinates into it coupled with something like google maps location history (https://maps.google.co.uk/locationhistory) again with everyone in the room adding screenshot of there location history is as hard to spoof as faking being in the hotel / venue. Especially if you couple this with the IP address / venue name, you have a lot of verification that you were there at the agreed time.

Thinking into the future I could see a crypto id secure photo camera that only switches on when it is in an agreed upon location at an agreed upon time for an agreed upon amount of time. These details would have to be previously put into the block chain. This could work very well as well.

patcon commented 10 years ago

Hm. Ok, this is interesting. Was thinking about this last night :)

For some reason, the approach of "proving" things with photos and such things doesn't sit comfortably with me. (Proving location with landmarks, proving trust and consensus via group photo, etc.) It seems very fuzzy...

I suppose I feel like when we don't have a tool in our toolkit yet to solve a problem, we should trust a person and use the tool of PGP until we have a proper tool. So if the location's representative and the passport issuer person (organizing the event) says we were here, then we were there. But maybe that's just me!

But, in terms of the looong game... :)

Mozilla Location Services (MLS) is building up a database of access points and cell towers: https://location.services.mozilla.com/

The idea is to build a database like Google has, but open, that can be used to triangulate location without GPS. So this is interesting. GPS is useless for us to verify location because GPS is read-only -- we just read it, and we don't send anything to it, so it can't corroborate. But wifi access points can potentially accept data, and be a third party to verify that you were there then, sending data to each nearby open access point....

EFF is working on something called the open wireless project: https://openwireless.org/

Their goal is to developer and proselytize wifi router firmware that is open source and progressively oriented. Their immediate goal is to push it out so that it, by default, opens a public and private access point on each router, throttled appropriately. The idea is that they want to create ubiquitous free wifi, and change the culture of sharing internet access. Maybe you see where I'm going with this.

So if we have a public record of all access points via MLS, and we can start depend on open access points, then we can route a set of location-verifying packets through these access points, and corroborate that we were located at a point between all the access points through which we routed packets.

Should EFF we willing to support such a use-case, then perhaps we could even have some sort of challenge-response sort of process, where the router can confirm the round trip data transfer time to the device. If you're not right there, then you won't be able to respond in an appropriate amount of time, so you must be actually there, and not routing your reponse through from a remote location.

MrChrisJ commented 10 years ago

Yeah this is good. It's something called point-of-presence. I like where you're taking this because then each node that connects to the open wireless it can act as a witness and make a claim as to its legitimacy.

Perhaps each wireless point could have its own bitcoin key pair generated inside of its executed code. Once executed it would be impossible for someone to find out the private key. Then this router would be an autonomous object, we could send it small bitcoin transactions and use OP_RETURN to issue commands live on the blockchain. It could return claims about its location in the form of its proximity to other wireless nodes within its line of sight. And then together they would all independently confirm their location in a mesh, proving a kind of internal consistency.