Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-05 Thread Andrew Cady
On Sun, May 04, 2014 at 10:51:40AM -0400, Nick wrote: Quoth Andrew Cady: On Sat, May 03, 2014 at 12:35:39PM -0400, Nick wrote: if you're worried about an evil google, hey, they control the browser, so you've already lost. I use Chromium and update it through my distro, so no, Google

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-04 Thread Griffin Boyce
Nathan Freitas wrote: Automated distributed deterministic build comparisons FTW! Seriously, it seems like we are pretty close with such a thing for Android APKs, so perhaps Chrome extension bundles could be added to the list, as well. That sounds pretty awesome :D Apps and extensions are

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-04 Thread Andrew Cady
On Sat, May 03, 2014 at 02:51:43PM -0400, Nathan Freitas wrote: On May 2, 2014 8:46:08 PM EDT, Griffin Boyce grif...@cryptolab.net wrote: On 2014-05-02 20:35, Andrew Cady wrote: On Fri, May 02, 2014 at 05:22:11PM -0400, Griffin Boyce wrote: I can't be vanned/rubber-hosed because

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-04 Thread Andrew Cady
On Sat, May 03, 2014 at 12:35:39PM -0400, Nick wrote: if you're worried about an evil google, hey, they control the browser, so you've already lost. I use Chromium and update it through my distro, so no, Google does not control the browser (/usr/bin/chromium). But they do, still, control the

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-04 Thread Nick
Quoth Andrew Cady: On Sat, May 03, 2014 at 12:35:39PM -0400, Nick wrote: if you're worried about an evil google, hey, they control the browser, so you've already lost. I use Chromium and update it through my distro, so no, Google does not control the browser (/usr/bin/chromium). Me too,

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-04 Thread Griffin Boyce
On 2014-05-04 01:02, Nick wrote: https://developer.chrome.com/extensions/crx is the documentation that mentions the signing. There are a couple of scripts there that will create a signed .crx file. I also wrote one a while ago[0]. I don't know how crx files integrate with Google's developer

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-03 Thread Tom Ritter
On 2 May 2014 17:22, Griffin Boyce grif...@cryptolab.net wrote: Do chrome extensions have a private offline key you use to sign extensions, to prevent malicious extension upgrades by google/an attacker who can middle SSL? No, though I have two-factor authentication using a secure device

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-03 Thread Nick
Quoth Tom Ritter: This makes it harder for someone to compromise your account, but not Google. In the Android App store, it's a *little* stronger, as apps are signed by a developer key, and they need that key to update. Except if Google really wanted they could push down an update to bypass

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-03 Thread Griffin Boyce
Nick wrote: Can you definitely not sign extensions with a private key? This is not an option available to any of my extensions or apps, unfortunately. There's reference to it in the documentation, but I've never seen this as an option for apps or for my developer account. Could you

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-03 Thread Nathan Freitas
On May 2, 2014 8:46:08 PM EDT, Griffin Boyce grif...@cryptolab.net wrote: On 2014-05-02 20:35, Andrew Cady wrote: On Fri, May 02, 2014 at 05:22:11PM -0400, Griffin Boyce wrote: No, though I have two-factor authentication using a secure device (not a cell phone), and I can't be

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-03 Thread Nick
Quoth Griffin Boyce: Nick wrote: Can you definitely not sign extensions with a private key? This is not an option available to any of my extensions or apps, unfortunately. There's reference to it in the documentation, but I've never seen this as an option for apps or for my developer

[liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-02 Thread Griffin Boyce
Hey all, So lately I've been obsessively working on a project to get software into people's hands and make it easy for them to see whether it's been tampered with in-transit. Code: https://github.com/glamrock/satori (download the zip) App:

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-02 Thread Tom Ritter
On 2 May 2014 11:00, Griffin Boyce grif...@cryptolab.net wrote: Also open to ideas about how I'm screwing this all up or am failing to account for Threat Model X. I'm wondering about the update mechanism. As I understand it, some scenarios are: 1) You bake in SHA256 hashes of software, with

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-02 Thread Griffin Boyce
Tom Ritter wrote: I'm wondering about the update mechanism. Do chrome extensions update over SSL? Is this update connection to google pinned, so you have to compromise a specific CA, instead of any CA? Chrome packaged apps update over SSL from a domain that has its certificate pinned.

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-02 Thread Andrew Cady
On Fri, May 02, 2014 at 05:22:11PM -0400, Griffin Boyce wrote: No, though I have two-factor authentication using a secure device (not a cell phone), and I can't be vanned/rubber-hosed because I don't actually know the password to my Google developer account. Some of this does require trust

Re: [liberationtech] Satori - distributed tamper-resistant circumvention tools

2014-05-02 Thread Griffin Boyce
On 2014-05-02 20:35, Andrew Cady wrote: On Fri, May 02, 2014 at 05:22:11PM -0400, Griffin Boyce wrote: No, though I have two-factor authentication using a secure device (not a cell phone), and I can't be vanned/rubber-hosed because I don't actually know the password to my Google developer