Gabo H Beaumont has a new note here:
Query Parameters Are Not Resource Identifiers
Using query parameters as the primary way to identify a document breaks with established web conventions. In the web architecture, the path component of a URL should identify the resource, while query parameters modify the request or representation.
This document seems to be pushing back against ideas introduced here:
I think we are in agreement that we should not use query parameters to identify resources. So what is a resource? Currently in our system a resource is a Document, Comment, Contact, Profile, or Capability.
A Document is a bunch of blobs that describe changes to the Document resource, but also you generally want to load everything who references a document so you can show backlinks and discussions. Generally you want to view these by constructing the document and looking at the content- that is the main view of this resource.
But you might want to view this resource data with a different presentation. For example you might want to see "whats new" on this document so therefore you'd get a feed view of it.
So in my opinion, a feed is not a resource. I believe you use a feed view to see "whats new" about the resource. That seems to be the fundamental thing that we need to resolve