Lars Yencken

From discussion: we could also consider a gradual migration, e.g. enabling pyarrow types for new datasets rather than applying it over everything. We could also bench the performance gains to...

Pablo noted that adding commitments to our future selves can mean that "small" data updates can expand. We should try to make more things as automatic as possible.

Thanks for your thoughts @Marigold! I totally understand that it can feel not valuable to invest in something people aren't using. The motivation for this was not about where we...

We can park this one for now, the type system is not the main focus in Grapher-land in the near future.

➡️ We agreed to discuss it in the next Monday data meeting

We said "nice to have" since we do have a system right now based on automatic Github issues, although it has limitations. @edomt Feel free to bump this up in...

Related to #2465

As of Pandas 2.0, you can now choose to use `pyarrow` dtypes. Apache Arrow uses a more consistent missing value scheme for floating point numbers, so these new dtypes do...

Pablo's intuition is that longer steps are more likely to break. Things we could try: - [x] Moar retries - [ ] A newer version of botocore - [ ]...

Suggest to pair with @Marigold early next week to try to resolve it.