Raw

You write: "...and do a GET on http://example.org/people#joe".



You can't. All you can do is a GET on "http://example.org/people". There's no place for the fragment identifier in the HTTP request.



This gets very important when we get to statements like "An RDF document retrievable at http://example.org/people#joe might not mention the resource http://example.org/people#joe. But it's still (by WebArch definition) a representation of that resource."



Absolutely not. You do a GET on "http://example.org/people" and what you get back (if anything) is a representation of the resource http://example.org/people. This may or may not have anything to do with http://example.org/people#joe (though it would be a bit silly if it didn't have at least some relevance).



"Yes, why not - a photo of Joe on his school trip to Fountains Abbey is still a representation of Joe,". Not in the web arch sense of the word "representation". If you do a GET and get back a bunch of bits with a media type image/jpeg then you've, presumably, got a *representation of a photograph* which depicts Joe. Joe is a person, not an information resource, so can't have a representation in the, admittedly rather limited, sense that web architecture uses the term.@en