Server-based aggregations

[sweco-semtne]

Would it be feasible to implement something similar or integrate with VegaFusion. It is a serverside accelerator for large datasets visualized with Vega

https://github.com/vegafusion/vegafusion/

[josevalim]

Yes, although I don't think we need a separate library or to use VegaFusion. Instead, using the Table.Reader protocol, we can ask the underlying table if it provides efficient aggregations. If it does, we can ask the underlying data structure to aggregate for us. This way we can automatically aggregate for you when using stuff like Explorer or other data structures.