ODD 26.? Objective - Harmonizing Earth Observation patterns across languages and GIS services

Michael Sumner shared this today on the Pangeo Discourse:

as a prolific R and GDAL user I would like to see a focus on cross-language consolidation. I think python is currently stuck in a deadend with rasterio, as good as that is it's not GDAL and it's not going to track future developement and it's not going to change substantially from its original (2011) design, xarray relies on a fourth party to interact with GDAL (rioxarray), and that itself has another alternative odc. I like python but it doesn't sit in an elevated position for me, R and python are the same in terms of the level of language they are but each are ahead or behind of what the other is good at in complex ways (and of course then there's Julia and javascript and surely others). I think it's essential that xarray embrace the multidimensional model of GDAL (I made a start with ai help in gh:mdsumner/gdx) and flush out that messy set of workarounds and plumbing. Or not use GDAL, nothing is irreplaceable. I have a lot to offer in this space as an experienced R dev and GDAL contributor, but sometimes I'm a bit cautious because my python exp is relatively low - I like it though and I'm getting better at it! In R we have seen significant changes, spatial was sp+rgdal, then raster made that much more useable, then that's forked into sf and terra (80% functionally equivalent), and then the rise of more general dev-focused packages: wk, geos, gdalraster, PROJ, arrow, sedonadb and many many others. For some tasks people were stymied by the "big 2 GIS packages", that were designed as comprehensive user-tools, at the expense of generality and flexiblity that devs need. rasterios is like that, GDAL itself is not and now has a very good mix of useability paired with its original hardcore abstract generality and API.

I know there's been work towards cross-language compatibility + GIS support on multiple VEDA teams, including science support and data services. I also think that ODD could help here by designing an objective that leverages this type of external community engagement. We could support a technical development group similar to our engagement with rasterix last spring.

cc @brianna-corremonte @abarciauskas-bgse @wildintellect

Start by reviewing the Pangeo Discourse post referenced in the issue and the discussion around cross-language GIS support. The issue mentions gdx, GDAL, rasterio, xarray, rioxarray, odc, and prior rasterix engagement, but names no repository files or tests. Done would require a defined objective and an agreed technical development or community-engagement scope.