Skip to content

Catechism - Discord Auditing, DAO use Case V0.0.1

Thoughts

I can go through every message and rate it if it is useful

Problem Statement

Find interesting and capable people on Discord Servers

What interactions am I personally looking for on Discord? 1

Mission Statement

Make digesting, mapping, and navigating the information inside discord servers easier.

What are you trying to do? Articulate your objectives using absolutely no jargon.

Discord is a powerful social media tool for managing communities. It's most used feature are Guilds that allow for groups of voice and text channels to be created with powerful RBAC(Rule Based Access Control) tools. It is common for Discord servers to contain 100's of people with 100's of thousands of messages, even millions. As an outsider joining a discord guild community it is difficult to understand what is going on. Many of the people communicating in the text channels know one another quite well therefore they know what one another is interested in, knowledgeable in, and how they like to be interacted with. All these text channels also often have multiple things being talked about all the time making them difficult to follow. I will be exporting discord servers into malleable data formats so the data can be played with to hopefully produce meaningful insights.

Just to make the task clear, I am going to develop tooling to generate context for Discord Guilds. The first task is to develop an analytics dashboard that can answer all the queries from Analysis Queries as outlined below. You can also check out Discord Binding to learn more about what these queries mean and do.

Analysis Queries Diagram

I have developed a dashboard like this previously and learned that it does not produce meaningful insights by itself. The purpose of the queries on the data is to produce insights that can lead to action, not just pretty graphs. I need a framework that can produce intelligence. For example,

  • We need to be able to separate the Admin announcement accounts from the actual users.
  • We need to be able to know who are doing the work to answer people's questions and who are asking the questions.
  • We need to know who is technical and who is not, plus what kind of technical they are.
  • We need to know how active a user is, for example do they post a question one a month and not even reply to the answer or do they ask a question every other day and give feedback on the answers in the channel.
  • We need to know if people are just talking into a void with people not responding.

All these insights by themselves do not produce intelligence. Once you know who the active people are within a discord community how are you supposed to engage with them? We need to annotate what they want. Know how they are trying to shape the world.

I first want to generate PDF reports from Jupyter from the basic insights then post them in the Discord Server highlighting who the important people are and how they are doing.

I want to have engaging persona's on all these DAO Discord Communities

I observe the usability problems with Discord being because the data layer is one dimensional. You can not annotate on top of content you can only reply or react. I want a tool like Twitter's Community Notes but available to the massed to annotate everyone elses content, with those same powerful RBAC tools.

There are many DAO's out there running on Blockchains. These DAO's coordinate using a variety of tools. One tool almost all DAO's have in common is Discord. I want to export all these Discord Server's so I can play with the data.

I want to produce reports that will be used to provoke these communities.

How is it done today, and what are the limits of current practice?

One can join a DAO's Discord server and see all the different channels, select each one, and scroll through it. The context switching within these channels is hard to keep track of. You have people coming in leaving a couple messages then completely other people wanting to talk about something else. The memory of the people conversing only goes back a couple days and everything before that basically does not exist.

What is new in your approach and why do you think it will be successful?

I will develop tools to track people as they navigate between channels on a single server.

I will develop tools to segment what is being talked about using tags and custom descriptions.

I will develop tools to perform the queries outlined in Analysis Queries.

Who cares? If you are successful, what difference will it make?

I will have access to information which will allow me to easily judge communities and the individuals that make them up.

What are the risks?

I waste my time.

I publicise this work all wrong and tarnish my reputation.

How much will it cost?

Lots of Time.

How long will it take?

  • I should have tiers of DAO auditing
    • 30 min
      • Social Media Audit with Comments
        • Discord
        • Twitter
        • Telegram
        • Github
    • 90 min
      • Contextualize Social Media Audit with types of messages
    • 4 Hours

What are the mid-term and final “exams” to check for success?

  • Discord Scraping Tutorial followed by someone else
  • Jupyter Notebook
  • Blog Post with analysis from top 10 DAO's
  • Hosted Dashboard
  • Graph Database Interrorgation