Create a machine image or Docker container to deploy set up environment?

[aSullivan-geo]

Feedback from customers about not having the admin rights needed to set up all the pieces needed for the toolbox to work, make it difficult for user to adopt. Specifically the conda environment, Google Cloud SDK etc.

Could we create a Google Machine Image that had all of the pieces needed already configured? Such that a user would deploy the machine image as a windows VM, and just need to sign into their ArcGIS online account for the license and provide the Google Cloud project with GEE enabled?

If we can, is there a place that we could provide that, either in GitHub or through GCP?

Also if there is a way to have a docker container that would do something similar that would probably work.

@schwehr and @KMarkert any thoughts?

[schwehr]

I don't know about including ESRI binaries / code in something like this. If it were all open source stuff, then that would be a reasonable thing to do and lots of projects do that.

I'm pretty sure geemap does that for docker:

• https://github.com/gee-community/geemap/blob/master/Dockerfile
• https://github.com/gee-community/geemap/blob/master/.github/workflows/docker-publish.yml