Add your questions or discussion directly on this page using the "Edit" button, or clone the repository as needed. .
Question: What is needed to ensure that a URI resolves differently when on the internal company network vs external (public) browsing?
The de-refernced public URI must only provide publicly available information, while the internal deferencing yields company confidential information in addition to the public content.
Ideally, a single URI deferences differently based on context (internal vs external).
Answer:
Redirects for resources served over HTTP/1.1 for linked data purposes typically use the 303 See Other code. These sorts of redirects (used by w3id.org and others) can only specify a single URI that the client should follow (as far as I can tell).
Thus, the implementation of internal vs external would need to be implemented by whatever server is listed in the redirect using something like a rewrite condition that has a clause to redirect the IP address of clients on the private network to an internal URI.
For example, if we have the URI http://w3id.org/some-pharma and someone visits this in a Web browser, the browser will receive a 303 See Other redirect that points to http://www.some-pharma.com. Then the server at http://www.some-pharma.com will check the IP address of the browser, using said rewrite condition, to see if they are connected to the internal network/vpn. If the browser is connecting from an IP that is internal to the company (i.e., on the network/vpn), then they are redirected to http://intranet.some-pharma.com where intenal content is displayed, else they simply see the public site.
So, I'm pretty sure that the implementation of this external/internal redirection rests on whomever owns the server at the initial 303 redirect. At least that is my initial take on it.