Skip to content

Catechism - CGFS Meme Model

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

Humans use a variety of social media applications such as Instagram, discord, reddit, twitter and the like. The foundational user experience across all these applications is very similar, users with identities sharing (posting) media in the form of text, images, and videos with rule based access control and the capacity to link to said media in interesting ways such as replies, likes, comments, reposts, etc, etc. At the end of the day all media is just files therefore we should be able to model the data from all these platforms using a regular data structure.

I want access to the raw data in whatever format I want so I can do analytics on it. I want to do RAG on all my social media. I also want to do RAG on my friends social media.

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

There exist apps such as Beeper, All in One Messenger, and grayjay that create a multi platform social media interface. There even exists Matterbridge which can bridge chat channels across a multitude of platforms.

The issue with the interfaces is that they are just that. I can't feed an interface into an embeddings model to perform RAG on. I need to access the underlying data structure.

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

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

What are the risks?

How much will it cost, time is a resource?

How long will it take?

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