you can't downvote replies so it wasn't him

I think xkcd should fund it, that would be comical.

Similar to working with an open source tool. Fix all the errors you find in the documentation when you are getting it up and running. If the pull request gets handled well and quickly, future contributions will probably be painless and productive.

How did you find $4 on Hetzner? I didn't see anything close!

Seems weird to call it sharding since it's not sharding indexed datasets or anything like that. Is this just a tool to mitigate Databricks’ internal service-scaling challenges?