Build on Briar instead of Ricochet #178

New Issue

Thomas_Leske · 2018-12-22T18:31:31Z

Thomas_Leske commented

2018-12-22 18:31:31 +00:00

Briar (https://briarproject.org/) can also use the TOR network for message transfer. However its protocol is already designed for asynchronous communication. And it provides serverless group communication.

Messages are end-to-end encrypted in a way that makes them appear random to anyone except for the intended recipiend. He is even able to fetch his message out of a stream of encrypted messages without public meta data on where a message begins and end (e. g. from an SD card after communication broke down).

I think one only has to integrate Cwtch servers to Briar in order to implement the features that the Cwtch project wants to provide.

Briar also passed an external security audit while Ricochet hasn't, yet.

Briar is only available for Android. However they are working on a web frontend that can be hosted on mainstream OSes.

Briar (https://briarproject.org/) can also use the TOR network for message transfer. However its protocol is already designed for asynchronous communication. And it provides serverless group communication. Messages are end-to-end encrypted in a way that makes them appear random to anyone except for the intended recipiend. He is even able to fetch his message out of a stream of encrypted messages without public meta data on where a message begins and end (e. g. from an SD card after communication broke down). I think one only has to integrate Cwtch servers to Briar in order to implement the features that the Cwtch project wants to provide. Briar also passed an external security audit while Ricochet hasn't, yet. Briar is only available for Android. However they are working on a web frontend that can be hosted on mainstream OSes.

Thomas_Leske commented

2018-12-22 18:45:16 +00:00

Offline message delivery is also a feature request on Briar:

https://code.briarproject.org/briar/briar/issues/1011

Offline message delivery is also a feature request on Briar: https://code.briarproject.org/briar/briar/issues/1011

sarah commented

2018-12-22 18:53:47 +00:00

I have had discussions with the Briar team on many occasions, and like Briar very much, however Briar and Cwtch have very different design goals. (https://fieldnotes.resistant.tech/metadata-resistant-consensus/)

I think of Briar as a tool for doing metadata resistant communication in environments where most participants already know each other (especially when existing infrastructure is unsuitable for comms). Cwtch on the other hand is being designed for environments where participants do not have prior relationships with each other but perhaps might want to get to know each other (with the option of falling back to anonymity.

Ricochet has also gone through an external security audit - https://ricochet.im/files/ricochet-ncc-audit-2016-01.pdf (although the primary metadata resistant mechanisms involved with Cwtch require independent review in any case, regardless of base)

In past discussions with the Briar team, their vision of offline message delivery is similar to Pond and very unlike Cwtch. Both approaches have merit and tradeoffs and I think it would be a disservice to both to shoehorn one into the other.

Ultimately, when we started Cwtch we reviewed a lot of existing approaches (including Briar) and technology (much of that discussion can be found in the whitepaper), and Cwtch originated as an approach that was separate from others and has evolved from there with input from the marginalized communities who we intend it to primarily serve.

Nothing from those inputs or from other discussions has made us reconsider the fundamental base of Cwtch.

I have had discussions with the Briar team on many occasions, and like Briar very much, however Briar and Cwtch have very different design goals. (https://fieldnotes.resistant.tech/metadata-resistant-consensus/) I think of Briar as a tool for doing metadata resistant communication in environments where most participants already know each other (especially when existing infrastructure is unsuitable for comms). Cwtch on the other hand is being designed for environments where participants do not have prior relationships with each other but perhaps might want to get to know each other (with the option of falling back to anonymity. Ricochet has also gone through an external security audit - https://ricochet.im/files/ricochet-ncc-audit-2016-01.pdf (although the primary metadata resistant mechanisms involved with Cwtch require independent review in any case, regardless of base) In past discussions with the Briar team, their vision of offline message delivery is similar to Pond and very unlike Cwtch. Both approaches have merit and tradeoffs and I think it would be a disservice to both to shoehorn one into the other. Ultimately, when we started Cwtch we reviewed a lot of existing approaches (including Briar) and technology (much of that discussion can be found in the whitepaper), and Cwtch originated as an approach that was separate from others and has evolved from there with input from the marginalized communities who we intend it to primarily serve. Nothing from those inputs or from other discussions has made us reconsider the fundamental base of Cwtch.

sarah closed this issue

2018-12-22 18:53:47 +00:00

Sign in to join this conversation.