Hey list,
have you already heard about the new hot shit?
Briar[1] is a "next level" messenger that is supposed to work without a central server and that wants to protect your meta-data through communicating via the tor network.
What do you think? Is the project going to be successful? When you follow the argumentation of Moxie Marlinspike [2], then it's probably not.
I hope that Briar is going to become popular and I'm going to download it as soon as it's available in the F-Droid store.
Greetings,
-- egnun
----- [1] https://blog.grobox.de/2016/briar-next-step-of-the-crypto-messenger-evolutio... [2] https://whispersystems.org/blog/the-ecosystem-is-moving/
On Tue, May 24, 2016 at 8:06 AM, Erik 'egnun' Grun egnun@gmx.de wrote:
What do you think? Is the project going to be successful? When you follow the argumentation of Moxie Marlinspike [2], then it's probably not.
The answer to Moxie's argument about federation is not centralization. That's a step back. The answer is distributed.
On Tue, 2016-05-24 at 11:47 +0300, Nikos Roussos wrote:
On Tue, May 24, 2016 at 8:06 AM, Erik 'egnun' Grun egnun@gmx.de wrote:
What do you think? Is the project going to be successful? When you follow the argumentation of Moxie Marlinspike [2], then it's probably not.
The answer to Moxie's argument about federation is not centralization. That's a step back. The answer is distributed.
Can you define what is distributed in this context ? Simo.
On Tue, May 24, 2016 at 10:11 PM, Simo simo@samba.org wrote:
On Tue, 2016-05-24 at 11:47 +0300, Nikos Roussos wrote:
On Tue, May 24, 2016 at 8:06 AM, Erik 'egnun' Grun egnun@gmx.de wrote:
What do you think? Is the project going to be successful? When you follow the argumentation of Moxie Marlinspike [2], then it's probably not.
The answer to Moxie's argument about federation is not centralization. That's a step back. The answer is distributed.
Can you define what is distributed in this context ?
Distributed means no servers at all. The same way P2P protocols work all these years in other fields than communication. You can read Torsten's post for more: https://blog.grobox.de/2016/briar-next-step-of-the-crypto-messenger-evolutio...
We have many efforts on this direction, besides Briar [1] mentioned above. Lately I'm using Richochet [2], but unfortunately is desktop only for now.
On 05/24/2016 07:06 AM, Erik 'egnun' Grun wrote:
have you already heard about the new hot shit? Briar[1] is a "next level" messenger
Erik! Hot shit?! The briar project started in 2011!
It recently gained another push by finally attracting some funding. I hope Michael Rogers, the guy behind it, was successful in recruiting good developers.
The main feature of Briar is that it is supposed to work in a purely peer-to-peer fashion without central servers, specifically in situations where there *is* no Internet connectivity (using Bluetooth, local meshes, etc). Tor is just the fallback mode for "online". In theory, Signal (and Whatsapp) could well use Tor to connect to its servers and provide some metadata protection too...
I think it is a very specialized solution for risk situations, and not (mean to be) a replacement for popular messengers. I tried the private beta back years ago and it still needed a lot of work. I do like the idea, and I think services that work when gateways to larger networks are cut will become more and more important...
It would be interesting to read a more thorough comparison to other similar products, which is not Telegram or Whatsapp or Signal, but things like the Serval project, also aimed towards peer-to-peer in mesh situations
On Tuesday 24 May 2016 18:05:19 Moritz Bartl wrote:
In theory, Signal (and Whatsapp) could well use Tor to connect to its servers and provide some metadata protection too...
That wouldn't help as the users could still be easily identified by their phone numbers. The server knows who talks to who how much.
Even if you use XMPP via Tor, the servers still know which JID talks to which JID how much. Logging in to your XMPP account once without Tor potentially compromises your identity to the XMPP server.
Kind Regards, Torsten
On Tuesday 24 May 2016 18:26:22 Moritz Bartl wrote:
Tat is incorrect. It would help.
You are right, it doesn't solve the problem, but it makes the problem smaller.
There are no absolutes.
Yes, I also assume that the NSA and friends have meanwhile acquired the capability to deanonymize at least part of the Tor traffic. Under this assumption, the metadata problem is still there.
Kind Regards, Torsten