[DOC] - Explain some Nebari concepts and use cases

Preliminary Checks

• This issue is not a question, feature request, RFC, or anything other than a bug report. Please post those things in GitHub Discussions: https://github.com/nebari-dev/nebari/discussions

Summary

Originally posted in https://github.com/nebari-dev/nebari/pull/1657

Who is this for, what problem does it solve? I know what JupyterHub does (I think). I have some pre-concieved notion of what dask does, what does Nebari do?

If defining exactly what Nebari does is too big an ask. Describe the workflow that it makes easier. maybe it helps with "Remember how different developing locally then deploying to the cloud is? Nebari makes that a thing of the past, All of your models run the same locally as in the cloud because you only run in the cloud"

How does OpenHPC fit in with Dask?

I get what kubernetes does, but at least while reading the README, the fact that this runs on top of kubernetes just tells me that this should be slightly easier to deploy. I still don't know why I would want to deploy this product.

Steps to Resolve this Issue

@iameskild suggested:

Hi @paddymul, welcome and thank you for raising these questions. Given how many different flavors of JupyterHubs there exist out there, I can definitely see how Nebari's existence can seem a bit confusing.

I'm all in favor of updating the README and our docs with more vivid use cases for Nebari compared to vanilla JupyterHub (and other flavors).

Let me see if I can quickly give a few high-level reasons why someone might want to use Nebari.

Deploying JupyterHub in the cloud can be tricky and requires a decent understanding of cloud infrastructure, kubernetes, and helm. Instead, Nebari can deploy an opinionated version of JupyterHub to one of several cloud providers (AWS, GCP, Azure, Digital Ocean) with a few simple commands (nebari init ... and nebari deploy ...).
Not only does Nebari deploy JupyterHub but it also integrates several very useful tools and features that most data scientist / data analysts and ML engineers use on a day-to-day basis. A few of these integrations include:
conda-store for environment management
dask to quickly and easily scale your code
Argo-Workflows for workflow management
Nebari comes with Keycloak for identity and access management, enabling greater control and flexibility over user administration.
Nebari can be deployed via GitOps making the management of the platform as simple as updating the nebari-config.yaml and opening a PR.
These are just a few of the reasons why someone might choose to use Nebari instead of rolling their own JupyterHub.

Again, I agree that we should be more explicit about the reasons why someone should choose Nebari over the alternatives.

Thanks again for bringing this to our attention and keep an eye out for an updated README ;)

Start by reviewing the README and existing Nebari documentation, especially the nebari init and nebari deploy workflow and nebari-config.yaml references. Identify where to explain Nebari's audience, use cases, integrations, and differences from vanilla JupyterHub. Done means a newcomer can understand why Nebari is useful and how its cloud, Kubernetes, Dask, and related concepts fit together.