While I am not someone who does much work in tradition web back-ends, this is idea I find interesting.

I talk a little bit about what an event store is and how that differs from a traditional database.

This involves discussions of some scalability and versatility problems which come about with traditional databases. From there, I talk a bit about how moving the source of truth to an event store can solve these problems.

I then go on to talk about the logic-level limitations to traditional event stores which block them being useful as this critical path source of truth, in general, and the shortcomings of some of the approaches used to address these.

Based on these point, I propose programmable event stores as a potential approach. This is essentially a combination of the approaches from programmable blockchains and event stores. Laminar is the example I produced to demonstrate this idea. I talk a bit about some of the observations I made while working on this project while walking through an example of how it could solve a simple problem. It does seem like a viable solution.

(recorded a few weeks ago)