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/650#issuecomment-218358483,https://api.github.com/repos/pydata/xarray/issues/650,218358483,MDEyOklzc3VlQ29tbWVudDIxODM1ODQ4Mw==,2443309,2016-05-11T04:25:04Z,2016-05-11T04:25:04Z,MEMBER,"@MaximilianR - I really like this idea.  I'm going to close this PR and we can continue to discuss this feature in the original issue (https://github.com/pydata/xarray/issues/422#issuecomment-218358372).
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,115933483
https://github.com/pydata/xarray/pull/650#issuecomment-185542959,https://api.github.com/repos/pydata/xarray/issues/650,185542959,MDEyOklzc3VlQ29tbWVudDE4NTU0Mjk1OQ==,2443309,2016-02-18T05:00:58Z,2016-02-18T05:00:58Z,MEMBER,"I'm doing some cleanup on my outstanding issues/PRs.  After thinking about this again, I'm not all than keen on pushing this into the `mean` method. I actually think it will end up being a bit of an ordeal to make happen.  `mean` is currently injected as one of the `NAN_REDUCE_METHODS`.  Its not entirely clear to me if it will be ""cleaner"" to refactor `mean` to support `weights`.  Thoughts?
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,115933483
https://github.com/pydata/xarray/pull/650#issuecomment-156139264,https://api.github.com/repos/pydata/xarray/issues/650,156139264,MDEyOklzc3VlQ29tbWVudDE1NjEzOTI2NA==,2443309,2015-11-12T15:32:07Z,2015-11-12T15:32:35Z,MEMBER,"Okay, let's go with the `mean` refactor.  We'll drop the `returned` arg and just add `weights` to the method.

@mathause - any comment?
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,115933483
https://github.com/pydata/xarray/pull/650#issuecomment-155478982,https://api.github.com/repos/pydata/xarray/issues/650,155478982,MDEyOklzc3VlQ29tbWVudDE1NTQ3ODk4Mg==,2443309,2015-11-10T16:34:41Z,2015-11-10T16:34:41Z,MEMBER,"> Any thoughts on the tradeoff between adding `average` vs adding a `weights` argument to `mean`? I guess it's nice that this mirrors NumPy.

That would be the main motivation.

If Pandas is going the way of pydata/pandas#10030 via mean, I think we could do that as well. I actually like that approach more since we tend to call it a ""weighted mean"" (see title of pandas issue).
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,115933483
https://github.com/pydata/xarray/pull/650#issuecomment-155161359,https://api.github.com/repos/pydata/xarray/issues/650,155161359,MDEyOklzc3VlQ29tbWVudDE1NTE2MTM1OQ==,2443309,2015-11-09T19:16:29Z,2015-11-09T19:16:29Z,MEMBER,"Thanks @maximilianr.  There has been an open issue here on this for a while (#422).

@shoyer  - I'm actually not sure I love how I implemented this but I'm teaching a session on open source contributions and code review today so I threw this up here as an example.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,115933483