Confronting Briar with disasters

Es gibt auch eine Übersetzung auf Deutsch

Unlike popular messaging apps Briar doesn’t rely on the Internet in order to keep working. Due to its peer-to-peer nature it can make use of various so-called “transports” to send messages even if the Internet connection is cut. Those transports can be virtually anything that you can use to send 0s and 1s, though currently only Bluetooth and Wi-fi are supported beside the Internet.

Briar was designed for journalists, human rights and environmental activists working in regions affected by surveillance and frequent Internet shutdowns. It’s therefore one of the very few messaging apps that not only try to protect the content of your messages but also the metadata attached to them.

TODO

Good luck if wildfires like these hit your radio towers (photo by Mike Lewelling, National Park Service, taken in 2013 of the Alder Fire in Yellowstone National Park, CC0. source)

Another potential use case scenario for Briar are natural disasters. With the climate crisis getting worse day by day, destruction of critical infrastructure is a problem affecting more and more parts of the world, as the recent floods in Europe and China and the wildfires all around the world have shown.

While Briar can definitively be useful in those situations, its trade-offs in favor of privacy are severely limiting its connectivity capabilities. To make an example, imagine your city just got nearly extinguished by a wildfire, destroying all the telecommunications infrastructure that was once there. Fortunately, you and your friends got Briar installed, so when a friend of you drops by you grasp at the chance and write messages to all your friends in-town. One could think that all those messages get synchronized to your friend’s device, so she can serve as a carrier for your other friends’ messages. Unfortunately, that’s not how Briar works.

As I’ve outlined before, metadata protection is one of Briar’s primary goals. Therefore, Briar doesn’t synchronize messages to your friend Alice with Bob when you meet him in order to not let Bob know that you’re communicating with Alice. This is very useful when you can’t trust even your contacts not to be spying on you, but it’s most likely a huge problem when connectivity is all you want in the face of natural disasters.

This message routing scheme used by Briar is called “single-hop social mesh” because you only ever send messages to your contacts if you have a direct connection to them. During catastrophes you most likely want to have at least “multi-hop social mesh” or yet even better “public mesh” where you share messages not only with your contacts but with anybody using Briar. However, as connectivity improves, privacy gets worse because people will know when you’re communicating with whom.

The good news are that Briar is currently receiving funding to conduct research on supporting other types of mesh. Still it will take a lot of time until something gets implemented in Briar, so all of this should be considered long-term perspectives. Note, though, that this mainly affects private chats and private groups. If you and all your friends are part of a forum (Briar’s “public” version of group chats), Bob will indeed serve as a carrier for your messages sent to that forum.

Another related problem when it comes to disasters is Briar’s ability to so far mostly only send text messages. True, recent releases of Briar introduced images, but those only work in private chats and are of bad quality due to heavy compression. Briar could offer a significant advantage over analogous walkie-talkie communication if it allowed to send high-quality voice, image and other attachments. There’s still a lot of work to be done on lower protocol levels of Briar to enable those new forms of attachments, but just like with mesh Briar already offers a solid basis that could be used to build a messenger focused on disaster communication.