Avalanche Design Proposal

Proposal

We’ll start with a secure messaging platform, standing on the shoulders of Signal. On top of that, we’ll build a powerful organizer toolkit, enabling organizers to build whatever they need to activate and nurture their own communities:

Lincoln has started working on it here: https://github.com/lincolnq/avalanche

Why this might be worth it

There are a lot of drawbacks to making people install a new app, but there are also huge potential upsides. We intend to overcome the drawbacks while providing access to the upsides.

Let’s take the example of a conference messaging tool. We want to enable anyone at a conference to find each other’s contact info by default. Conferences are high-investment events (people are committing a full day minimum or more) to the conference, and most “conference apps” (Swapcard, Whova, etc) kinda suck. One of the main ways they suck is that they’re not messaging first—messaging is an afterthought—and as a result it’s quite unreliable to actually get in touch with people on the platform.

If conferences were organized on Avalanche instead, they would:

The design centers on self-hosted Signal-quality encrypted messaging — a unified inbox of all your conversations across all your activism and associated socials — with a project platform to enable anyone to rapidly build organizing tools that are specifically directly integrated into the messaging.

The project platform will enable projects like:

The Avalanche platform is self-hosted and open source. Specifically, any group who wants to set up the platform controls their own “server”: think a Slack server or a Discord server, but even more under the group’s control because the platform is open source and (optionally) self hosted. Groups can set standards for who is invited to the server and what kinds of automatic access they have.

Similar to Slack or Discord, there will only be one shared Avalanche app to download, because getting people to download a new messaging app is quite difficult & costly — so trying to make the app as general-purpose as possible, in order to not need to solve this problem again in the future!

Project Plan

We can build this relatively quickly because Signal has open-sourced most if not all of their platform, and Claude can pick up from where Signal left off. We can use libsignal as the basis of our secure communication system, rebuild all messaging UX on top of it, and then add on the Project platform.

I (Lincoln) have been soloing this for the last several weeks using Claude, but I’ve only had time to work on it a few hours per week. I have gotten fairly far, but there’s still a lot to do.

← All posts