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/2891#issuecomment-482909157,https://api.github.com/repos/pydata/xarray/issues/2891,482909157,MDEyOklzc3VlQ29tbWVudDQ4MjkwOTE1Nw==,6355796,2019-04-14T01:10:56Z,2019-04-14T01:10:56Z,NONE,"Yes good point.  I do also think that the expand_dims interface should really only be responsible for that one single operation.  If you then want to make a copy then go ahead and use that separate method afterwards.

Ok great, if I get time later today I'll see if I can't pick that up; that is if someone hasn't already done so in the meantime.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,432798510
https://github.com/pydata/xarray/issues/2891#issuecomment-482907774,https://api.github.com/repos/pydata/xarray/issues/2891,482907774,MDEyOklzc3VlQ29tbWVudDQ4MjkwNzc3NA==,6355796,2019-04-14T00:43:35Z,2019-04-14T00:43:35Z,NONE,"Thank you @shoyer for taking the time to educate me on this.  I understand completely now.

I agree that solutions one and two would be helpful for future developers new to Xarray and the expand_dims operation when they eventually encounter this behaviour.  I also agree that option three would be confusing, and had it of been implemented as such I would still have found myself to be asking a similar question about why it is like that.

Another option to consider which might be easier still would be to just update the expand_dims documentation to include a note about this behaviour and the copy solution.

Thanks again!","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,432798510