html_url,issue_url,id,node_id,user,created_at,updated_at,author_association,body,reactions,performed_via_github_app,issue
https://github.com/pydata/xarray/issues/4156#issuecomment-844090889,https://api.github.com/repos/pydata/xarray/issues/4156,844090889,MDEyOklzc3VlQ29tbWVudDg0NDA5MDg4OQ==,2448579,2021-05-19T13:09:25Z,2021-05-19T13:09:25Z,MEMBER,"There is a more standards-compliant version here:https://github.com/pydata/xarray/issues/1077#issuecomment-644803374

This is still blocked on choosing which CF representation to use for sparse vs which one to use for MultiIndex.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,638947370
https://github.com/pydata/xarray/issues/4156#issuecomment-644417331,https://api.github.com/repos/pydata/xarray/issues/4156,644417331,MDEyOklzc3VlQ29tbWVudDY0NDQxNzMzMQ==,6815844,2020-06-15T22:13:50Z,2020-06-15T22:13:50Z,MEMBER,"Do we already have something similar *encoding* (and decoding) scheme to write (and read) data?
(does CFTime use a similar scheme?)
I think we don't have a scheme to save multiindex yet but need to manually convert by `reset_index`.
#1077

Maybe we can decide this encoding-decoding API before #1603.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,638947370
https://github.com/pydata/xarray/issues/4156#issuecomment-644372749,https://api.github.com/repos/pydata/xarray/issues/4156,644372749,MDEyOklzc3VlQ29tbWVudDY0NDM3Mjc0OQ==,2448579,2020-06-15T20:32:01Z,2020-06-15T20:32:01Z,MEMBER,"Yes I think we will have to ""encode"" to something like this example
```
dimensions:
  lat=73;
  lon=96;
  landpoint=2381;
  depth=4;
variables:
  int landpoint(landpoint);
    landpoint:compress=""lat lon"";
  float landsoilt(depth,landpoint);
    landsoilt:long_name=""soil temperature"";
    landsoilt:units=""K"";
  float depth(depth);
  float lat(lat);
  float lon(lon);
data:
  landpoint=363, 364, 365, ...;
```

and then write that ""encoded"" dataset to file.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,638947370
https://github.com/pydata/xarray/issues/4156#issuecomment-644368878,https://api.github.com/repos/pydata/xarray/issues/4156,644368878,MDEyOklzc3VlQ29tbWVudDY0NDM2ODg3OA==,6815844,2020-06-15T20:27:37Z,2020-06-15T20:27:37Z,MEMBER,"@dcherian 
Though I have no experience with this gather compression, it looks that python-netcdf4 does not have this function impremented.

One thing we can do is
`sparse -> multiindex -> reset_index > netCDF`
or maybe we can even add a function to skip constructing a multiindex but just make flattened index arrays from a sparse array.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,638947370