Why use S3 for download, and how to download if S3 is unavailable

[ryancwalsh]

I'm adding this issue from https://discord.com/channels/490367152054992913/542945453533036544/996106267263238225 per George M's request.

---

I'm now following instructions (https://near-nodes.io/validator/compile-and-run-a-node#5-get-data-backup-1) to run a mainnet node locally because of https://stackoverflow.com/a/72131773/470749

I was surprised to see that downloading from AWS S3 is a step.

That doesn't sound decentralized. Why wouldn't it be possible to spin up a node without a centralized dependency? Thanks.

You can spin up a node from scratch but it would take FOREVER to sync and catch up to the current block. The backup is there to speed up the process AFAIK

Ok, thanks. Who would know for sure, and could we update the docs to show the steps for not relying on S3 and for clarifying why we recommend the shortcut of using S3?

---

Maybe George M's team can clarify in the docs.

I do not need the answer urgently. Instead I'm submitting issues to docs for 3 reasons:

1. Out of curiosity, to increase my personal understanding
2. To improve the docs for others
3. More importantly, since we all want NEAR to accelerate like a rocket, we need to have systems and processes in place to be prepared to accept and benefit from lots of contributions to docs and tools. There will be a lot of people who want to help. So we'll want to find ways to encourage questions and suggested edits and handle the influx.