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/pull/5933#issuecomment-1016805677,https://api.github.com/repos/pydata/xarray/issues/5933,1016805677,IC_kwDOAMm_X848mzkt,39069044,2022-01-19T19:39:57Z,2022-01-19T19:39:57Z,CONTRIBUTOR,"Not sure I understand the blockwise approach well enough to make a PR, but maybe I'll give it a try at some point.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,1043746973
https://github.com/pydata/xarray/pull/5933#issuecomment-1015887354,https://api.github.com/repos/pydata/xarray/issues/5933,1015887354,IC_kwDOAMm_X848jTX6,39069044,2022-01-18T22:21:49Z,2022-01-18T22:45:16Z,CONTRIBUTOR,"> @slevang are you still interested to continue this PR? I think that would be a worthwhile addition and should not be too much left to do. (What would be nice, however, are tests for the issues this fixes.)

Definitely! I got distracted is all, and @dcherian posted a nice solution in [#5629](https://github.com/pydata/xarray/issues/5629#issuecomment-960133879) that could allow us to preserve the ability to fit along a chunked dimension using blockwise operations and the dask `lstsq` implementation used by the existing `polyfit` code. 

I'm happy to pick this back up and finish it off if there is consensus on the right way forward, but the blockwise approach seemed promising so I put this on hold.  

","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,1043746973