Question: Is expanding beyond S3 (e.g., IPFS/libp2p, SSH3, etc.) in scope for this project?

[diegoripley]

Hi,

I’ve been looking into this project and I’m interested in how extensible the storage backend model is.

From what I understand, the current implementation is primarily designed around S3-compatible object storage. I’m curious whether it would be considered in scope for this project to support additional transport or storage protocols beyond S3.

For example:

• IPFS / libp2p
• SSH3-based transports
• Other distributed or peer-to-peer storage systems

My thinking is that a pluggable backend abstraction could allow multistore to route requests not only to different S3 buckets or regions, but also to different storage protocols entirely, depending on configuration or routing rules.

I realize this could introduce significant architectural complexity, so my main question is:

• Is multi-protocol storage support something that aligns with the long-term goals of this project?
• Or is the design intentionally focused on S3-compatible object storage only?

I’d appreciate any insight into the intended scope or architectural constraints.

Thanks for building and maintaining this project.