Untitled Contact shared a few weeks back this repo about the Spanish laws and we believe this is a good opportunity to use Seed to bring collaboration to something as important as the Spanish Law
The main goal
Seed is a powerful tool to bring content like this, because we can track changes, collaborate and help shape complex problems.
Idea 1: is this repository as the baseline and import data from the repository
We can create a connector or similar that listens (either automatically or manually) to the repo changes and import all the changes to it into Seed documents
While this is good, the issue I see is that we are at the mercy of what the actual data extractor will do. If the data extractor from the original data source (BOE) to the git repo changes, we either need to adapt or it can break.
Idea 2: Fork the repo and instead of extract data to the repo, extract data directly to Seed documents
This involves understanding how the data extraction looks like, and instead of materializing the data extraction to markdown documents (current repo work), we can do it to Seed dcuments directly.
I like this option better, because I believe this is far more "composable" and we can use the same techniques to other industries or data sources.
Observations
While I believe we should avoid working with the repo, creating a Github connector to Seed sounds also very important. specially to maybe create documentation againts codebases. we can use LLMs to understand codebases, understand changes, read release notes, and anything we believe is relevant to actually create something useful in terms of documentation.
Do you like what you are reading?. Subscribe to receive updates.
Unsubscribe anytime