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/1475#issuecomment-457907278,https://api.github.com/repos/pydata/xarray/issues/1475,457907278,MDEyOklzc3VlQ29tbWVudDQ1NzkwNzI3OA==,1197350,2019-01-27T10:49:07Z,2019-01-27T10:49:19Z,MEMBER,"> I'm not sure I understand (N,M) sized coordinates for unstructured meshes -- what is M here? The total number of cells? Some arbitrary constant indicating the maximum number of sides for a single cell?

N is the number of cells. M is the number of points required to specify the cell vertices, e.g. 4 for 2D quadmesh, 3 for 2D trimesh, 8 for 3D quadmesh, etc.

Regarding your options 1 or 2, I guess I'm agnostic as to how it is implemented. I recognize 2 introduces lots of complications. What matters is how it will interact the indexes, i.e. can we easily select data based on cell bounds?

I will have to take some time to think about what you wrote, as it is hard for my brain... 🙃 ","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,242181620
https://github.com/pydata/xarray/issues/1475#issuecomment-457246424,https://api.github.com/repos/pydata/xarray/issues/1475,457246424,MDEyOklzc3VlQ29tbWVudDQ1NzI0NjQyNA==,1197350,2019-01-24T15:50:24Z,2019-01-24T15:50:24Z,MEMBER,"I'm just pinging this issue again to keep it fresh. 

I am becoming more and more convinced that we need to allow for cell bounds in xarray's data model. Contrary to my comments above, I no longer think this is a problem to be solved with xgcm or some outside package.

CF conventions, which we partially support in other parts of xarray, have a clearly defined concept of [cell geometry](http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/cf-conventions.html#_data_representative_of_cells). When present, such coordinates could decoded and used for indexing and plotting.

Currently we distinguish between ""dimension coordinates,"" which are converted to indexes, and ""non-dimension coordinates."" What if we added a new type of coordinate called ""cell coordinates""? We could accomodate either (N+1) sized coordinates for quad-mesh geometries of (N,M) sized coordinates for unstructured meshes.

What is a concrete first step we could take towards this goal? Try to work out a design document?

","{""total_count"": 6, ""+1"": 6, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,242181620
https://github.com/pydata/xarray/issues/1475#issuecomment-417172168,https://api.github.com/repos/pydata/xarray/issues/1475,417172168,MDEyOklzc3VlQ29tbWVudDQxNzE3MjE2OA==,1197350,2018-08-30T02:49:12Z,2018-08-30T02:49:12Z,MEMBER,"cc @adcroft, who expressed interest in this topic.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,242181620
https://github.com/pydata/xarray/issues/1475#issuecomment-314898489,https://api.github.com/repos/pydata/xarray/issues/1475,314898489,MDEyOklzc3VlQ29tbWVudDMxNDg5ODQ4OQ==,1197350,2017-07-12T21:12:09Z,2017-07-12T21:12:24Z,MEMBER,"These are precisely the sort of issues we are trying to solve with [xgcm](https://github.com/xgcm/xgcm). I am about to make a big new release. Using the xgcm concept of an `Axis` object (not yet in the online docs until the new release), it should be pretty easy to add this sort of plotting support in an arbitrary number of dimensions. ","{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,242181620