search

2i2c-org / infrastructure

Infrastructure for configuring and deploying our community JupyterHubs.
https://infrastructure.2i2c.org
BSD 3-Clause "New" or "Revised" License
103 stars 63 forks source link

Codify shared responsibility model 'lines' for the NASA VEDA Hubs #3962

Open yuvipanda opened 5 months ago

yuvipanda commented 5 months ago

Thanks to efforts from me and @colliand, we've been able to get more resources from Development Seed for working with the NASA VEDA, GHG and ESDIS hubs. They're a very ideologically aligned organization, but focused more on development rather than infrastructure (https://github.com/2i2c-org/leads/issues/260 has more information).

From the perspective of 2i2c engineering, this should result in a net decrease in the amount of work we have to do for these hubs. In particular, I think it would:

  1. Establish a 'first level of support' that isn't us. So fewer support requests coming our way.
  2. Move responsibility for image management completely away from us.
  3. Provide more engineering (and review) resources for upstream Jupyter development in various ways. Note that this doesn't commit 2i2c engineering resources to anything.
  4. They would also be able to help handle requests in many ways. In terms of https://github.com/2i2c-org/meta/issues/953#issuecomment-2030777639, a lot of 'feature enablement' or 'config change' will instead turn into 'community config change review'. https://github.com/2i2c-org/infrastructure/issues/3735 for example would be something off our hands.

They love the 2i2c shared responsibility model. This issue tracks clarifying these, so we know who is exactly responsible for what with the now increased capacity.

yuvipanda commented 5 months ago

There's now an official "NASA VEDA JupyterHub team lead", which happens to be @batpad (from Development Seed). He's contributed to kubespawner before, has a lot of tech lead expertise, and also is my very long time friend, initially via the open source world.