Rafael Schouten

I think `combine` is best because DataFrames uses it. Maybe it would be `summarise` in R.

See: https://github.com/JuliaData/DataAPI.jl/issues/65 for using `combine` without clashes with DataFrames.jl

Rasters uses `Base.read` and DD forwards it on to the parent (have to check that). Maybe YAX could add that too?

Probably as Zarr as well. Maybe we can just move Rasters backends here but ignore CF standards and other geo things

I've been thinking about this. We need a generalised and extensible writer backend for all AbstractDimArray. So people could add their own package that defines a DD compatible writer. What...

Yeah, we need to combine YAX and Rasters readers into a more general DimensionalData/DiskArrays reader/writer interface. I'm not sure where these things live, and if they would all be separate...

Ya I will do it soon! Just have a backlog of DD/Rasters PRs and no time

I think we need to be very organised about how we do this. We already have two DD packages that load data we dont want to have three! Rasters will...

Hard to say, but at least a month full time, maybe 2, to get both the packages using the same backend with a good design. (Maybe we should organise a...

No, it's not something I can chase as I have too much work currently. But if someone organised something I could find time to work on it in the next...