Deepak Cherian

👍🏾 for updating the docstring. IME `load` is commonly used so I don't know that it's worth removing.

`xr.broadcast(ds)[0].sum(dims)` should do this. We could add it here: https://xarray.pydata.org/en/latest/howdoi.html and to the docs under [Aggregations](https://docs.xarray.dev/en/stable/user-guide/computation.html#aggregation)

Yeah I think we need to fix this in flox. Can you come up with a simple test case that checks that the accumulation is done properly?

This is quite confusing and I doubt it's intentional. I would've expected `b (y) int32 3 3` assuming that it would've been broadcast along the reduction dimension. The final value...

We discussed: 1. dropping variables without the dimension 2. Return ds.sizes["x"] by broadcasting `b` along `x` ----- For the other reductions ``` python import numpy as np import xarray as...

Shall we close? It seems like the current code is what we want.

> his could be replaced by importlib.util.find_spec. Nice. Does it work on python 3.8? > However, many backends also check for ImportErrors (not ModuleNotFoundError) that occur when a library is...

Thanks for leading this effort @keewis. I would start with the lowest-level operations like the constructors and `align`, `concat`, `merge`. These are called in many of the other functions so...

@keewis Shall we close this? It seems the only outstanding one is `nanprod` with quantities. Which sort of indicates that we've made the necessary big changes.

> Given the performance behavior I'm guessing we may be doing sequntial interpolation for the dimensions, basically an interp1d call for all the xx points and from there another to...