Re: [liberationtech] Building a encrypted mobile network
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/06/13 17:47, Jonathan Wilkes wrote: Concealing these patterns would require users to send and receive dummy data even when they weren't sending or receiving calls, which would drain their batteries and data allowances. It would be possible to build such a system, but I don't think anyone would use it. I don't think it's out of the realm of possibility that somebody would have a device running orbot with a (non-exit) relay that sits at home, plugged in, running over wifi. Or, some small plug computer with a headset hookup that functions the same. Or on their main machine that just runs all the time. All that's needed then is a mechanism to leave a text message when the other person isn't at home (Torchat, maybe Bitmessage, etc.). Well yes, if you take the mobile out of mobile security, the problem gets easier. ;-) Seriously though, I agree that this could work really well on a Freedombox or similar. Cheers, Michael -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) iQEcBAEBAgAGBQJRuFRfAAoJEBEET9GfxSfMDTMH/35TjGWNl64DrZCvrWRmafO3 qMfufyA6dPPY+0rix7ptMOu4DSyMQ36Q05AdFM0HxB+c2O4wP9nHjSSFq8ba094D /NobvVBrg0Rhn0hNEJ5nMf4yJV1O7LkV+jhDLBJZS+1dYybwJX9LqMQxlBYJnqZG ykLxU0/fFG8XxAi+6fJjsbtO0gRAQqoaq4cByXa9FgtPnleXNaSPD+erGXGoKFIj 4Tbq8dEGOzSGhCK6KGxKn1QKwCxk38G/kxFlg1oZYrZgr3ePdr/5ch5x40by6tzn jv4IqYC6I33+FKc1vcu4eEK+lw89/t9sqt/togHky3j2vhheqV4xbU3uVzF7dv8= =aOCf -END PGP SIGNATURE- -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at compa...@stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
Re: [liberationtech] Building a encrypted mobile network
From: Michael Rogers mich...@briarproject.org To: Jonathan Wilkes jancs...@yahoo.com; liberationtech liberationtech@lists.stanford.edu Sent: Wednesday, June 12, 2013 6:58 AM Subject: Re: [liberationtech] Building a encrypted mobile network -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/06/13 17:47, Jonathan Wilkes wrote: Concealing these patterns would require users to send and receive dummy data even when they weren't sending or receiving calls, which would drain their batteries and data allowances. It would be possible to build such a system, but I don't think anyone would use it. I don't think it's out of the realm of possibility that somebody would have a device running orbot with a (non-exit) relay that sits at home, plugged in, running over wifi. Or, some small plug computer with a headset hookup that functions the same. Or on their main machine that just runs all the time. All that's needed then is a mechanism to leave a text message when the other person isn't at home (Torchat, maybe Bitmessage, etc.). Well yes, if you take the mobile out of mobile security, the problem gets easier. ;-) Seriously though, I agree that this could work really well on a Freedombox or similar. It could work well with Tor and a cross-platform gui toolkit that allows it to run on OSX, Windows, GNU/Linux, and (ideally) Android. But yes, if someone developed such an application and got it running on a freedom box[1] I agree that would be extremely useful. Because after all, the freedom box[1] is a widely popular, well-documented, well-supported, and (relatively) inexpensive piece of hardware used not just by computer experts, but also educators, children, entrepreneurs, hobbyists, activists... all kinds of people, all around the world who care about having control over their machines and their data. [1] www.raspberrypi.org -Jonathan -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at compa...@stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
Re: [liberationtech] Building a encrypted mobile network
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Anthony, On 08/06/13 13:36, Anthony Papillion wrote: 1. Location is a particularly thorny issue. Presentations at either HOPE or BlackHat demonstrated how easy it is to locate a mobile even if you're not the government with a massive budget and mad technology. Perhaps routing the network connection through Tor may suffice? But I don't think so as something doesn't 'feel' right about that. Thoughts? Routing the call through Tor wouldn't conceal the phone's location from the mobile network. The caller and callee would both have to use cell towers to reach the Tor network, so their respective mobile networks would still know their locations, and any hacks that can currently be used to trick the mobile network into revealing a phone's location would still work. In theory you could conceal who calls whom from the mobile network by routing the call through Tor. However, in order to be able to receive calls, the callee would either have to maintain a constant connection to Tor (draining her battery and data allowance) or ask some third party with a constant connection to Tor to send her push notifications of incoming calls, which she could then answer by connecting to Tor. The third party would know when the callee was receiving incoming calls, though not necessarily from whom. Even this would reveal quite a lot of information to the mobile network. Alice starts sending data at 12:34:56. Bob receives a push notification at 12:34:57. Bob starts sending data at 12:34:58. Alice and Bob both stop sending data at 12:44:58. The inference is pretty clear: Alice called Bob at 12:34 and the call lasted ten minutes. Concealing these patterns would require users to send and receive dummy data even when they weren't sending or receiving calls, which would drain their batteries and data allowances. It would be possible to build such a system, but I don't think anyone would use it. 2. Content is much easier to protect. My initial thought is to take a stock Android phone, replace the dialer with a SIP client capable of doing ZRTP, and customize the phone to tower communication so that all communication between the two is fully encrypted (and I don't mean the BS GSM encryption). Once the data gets on the network, it would be decrypted and calls would be connected. Content would be protected automatically when the user called ANY SIP device that supported ZRTP. Calls to PTSN would still be wide open. It's not practical to use a custom protocol between the phone and the tower - apart from the logistical issues of rolling out a new protocol, carriers won't adopt a protocol that lacks lawful intercept backdoors. However, phone-to-tower encryption isn't needed if you have phone-to-phone encryption, so I believe RedPhone does what you want (but I haven't used it so I could be wrong). Cheers, Michael -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) iQEcBAEBAgAGBQJRtzfsAAoJEBEET9GfxSfMcfoH/jPBVjyJCBKThYy/kN14ZcNX pwaOzHdpZ+MxHKo919Exu2XUn9nIHlrGB1sqL9azsxss+m/bTgfc9iXVrOXQLhNb 8fif2PYacKgZ7eyrV1lFYesDXbcpgrRkFI7qJodc3ukfgZx87pmHmogXRGGpVvGy cx7X/+tXBPqi84Sq2tDRcPdX7eDRXxjoE6DK0YG6f9+KN3aPLfoFCQZrnMUzqgcG 6zvJrpuCvSiH1Uk5UMbjDGMsXempFf5kDTbThOhYJG2Fi+kOw9cOlsFx0z2QB5Yf 0dSRrTHPYOIxA+JwI0pRxhCnEOC8SEWCmQVzpzEww8RvK2/k0x5ZFBERtetxiRg= =irF4 -END PGP SIGNATURE- -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at compa...@stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
Re: [liberationtech] Building a encrypted mobile network
From: Michael Rogers mich...@briarproject.org To: liberationtech liberationtech@lists.stanford.edu Sent: Tuesday, June 11, 2013 10:45 AM Subject: Re: [liberationtech] Building a encrypted mobile network -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Anthony, On 08/06/13 13:36, Anthony Papillion wrote: 1. Location is a particularly thorny issue. Presentations at either HOPE or BlackHat demonstrated how easy it is to locate a mobile even if you're not the government with a massive budget and mad technology. Perhaps routing the network connection through Tor may suffice? But I don't think so as something doesn't 'feel' right about that. Thoughts? Routing the call through Tor wouldn't conceal the phone's location from the mobile network. The caller and callee would both have to use cell towers to reach the Tor network, so their respective mobile networks would still know their locations, and any hacks that can currently be used to trick the mobile network into revealing a phone's location would still work. In theory you could conceal who calls whom from the mobile network by routing the call through Tor. However, in order to be able to receive calls, the callee would either have to maintain a constant connection to Tor (draining her battery and data allowance) or ask some third party with a constant connection to Tor to send her push notifications of incoming calls, which she could then answer by connecting to Tor. The third party would know when the callee was receiving incoming calls, though not necessarily from whom. Even this would reveal quite a lot of information to the mobile network. Alice starts sending data at 12:34:56. Bob receives a push notification at 12:34:57. Bob starts sending data at 12:34:58. Alice and Bob both stop sending data at 12:44:58. The inference is pretty clear: Alice called Bob at 12:34 and the call lasted ten minutes. Concealing these patterns would require users to send and receive dummy data even when they weren't sending or receiving calls, which would drain their batteries and data allowances. It would be possible to build such a system, but I don't think anyone would use it. I don't think it's out of the realm of possibility that somebody would have a device running orbot with a (non-exit) relay that sits at home, plugged in, running over wifi. Or, some small plug computer with a headset hookup that functions the same. Or on their main machine that just runs all the time. All that's needed then is a mechanism to leave a text message when the other person isn't at home (Torchat, maybe Bitmessage, etc.). It's reinventing old technology: the landline and the answering machine. But users would avoid the new surveillance problems with metadata leaking. Whoever is planning the Restore the Fourth Amendment project would certainly make use of such a system if it existed and was usable. -Jonathan -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at compa...@stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
[liberationtech] Building a encrypted mobile network
Hi Liberation Tech! With the NSA spying scandal in full swing, I've been thinking about what it would take to truly build a secure mobile network. I'm curious to get feedback from those who've given more thought to this than me as I see the problem as primarily twofold: 1. Location issues - they know WHERE you are. 2. Content issues - they know what you say and who you say it to. If the two issues above are the main (only?) issues hindering the creation of a secure network, how could we work around them? Some thoughts: 1. Location is a particularly thorny issue. Presentations at either HOPE or BlackHat demonstrated how easy it is to locate a mobile even if you're not the government with a massive budget and mad technology. Perhaps routing the network connection through Tor may suffice? But I don't think so as something doesn't 'feel' right about that. Thoughts? 2. Content is much easier to protect. My initial thought is to take a stock Android phone, replace the dialer with a SIP client capable of doing ZRTP, and customize the phone to tower communication so that all communication between the two is fully encrypted (and I don't mean the BS GSM encryption). Once the data gets on the network, it would be decrypted and calls would be connected. Content would be protected automatically when the user called ANY SIP device that supported ZRTP. Calls to PTSN would still be wide open. Is this workable in any form or fashion? Am I a complete babbling idiot? Is anyone working on this currently? Your thoughts are most welcomed. Regards, Anthony -- Anthony Papillion Phone: 1.918.533.9699 SIP: sip:cajuntec...@iptel.org iNum:+883510008360912 XMPP:cypherpun...@jit.si www.cajuntechie.org -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at compa...@stanford.edu or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech