120: Materialize Origins: A Timely Dataflow Story with Arjun Narayan and Frank McSherry
The Data Stack Show
00:00
The Uncanny Valley of Developer Tooling
There's a really uncanny valley here. It's very interesting for anyone who has worked in developer tooling and sort of, you know, I think it's even more profound when you're doing a, like, data infrastructure. There are literally thousands and thousands of lines of Python that is moving data around. And then there's a long list of, oh, yeah, but don't use aggregates other than some encounters. A lot of them are just like, don't join joints or do the primary keys on materialized view implementations. You have the same problem with databases and materialized view Implementations. We chose Postgres to implement our big data system because its caveats needed to be
Transcript
Play full episode
