Write blog post about Learnings from a Data Engineer while exploring the Open Data Ecosystem

[davidgasquez]

Write a blog post in the vein of "Here's what I learned as an data engineer from a month deep dive into the open source data and the open data ecosystem".

A "how could the future look like" section could mention:

• LLMs to reduce coordination friction (data-GPT to automate schema, types and location inference. point it somewhere, get a datapackage.json)
• Potential role of WASM and modern tooling to make "small data" larger.

[davidgasquez]

A bit more on the "How the Future could look like".

Played with Langchain + OpenAI a bit more. It is now easy to run arbitrary queries to a database with Natural Language queries (see the Querying a Database section I did here).

I was thinking... it should be possible to create a LangChain tool that can access all open data and query it. Two quick approaches.

1. Search for relevant datapackages related to the topic T in GitHub. Add them as external tables and query them.
2. Have something like SplitGraph that is already agregatting all datasets into a CDNed database and use that directly. :point_left: I'm going to try this one quickly with Splitgraph and report back!

cc @rufuspollock

[rufuspollock]

@davidgasquez would be cool to write this up (or if you have already to x-post it)

[davidgasquez]

I don't have it and agree it would be cool to put something out in this vein.

Any thoughts or points that you think it might be interesting to cover?

[rufuspollock]

@davidgasquez

• First, brief info your personal original motivation
• Second, as few or as many examples of what you found - whatever is easy to do (e.g. just start with one or two and then we can add more in subsequent posts or a wiki page)
• Three: what did you find out
• Four: any questions / suggestions you want to leave the reader with ...

I imagine as you draft stuff will come up - also feel free to just dump ideas / outline and then we can iterate from there 😄