RedundantUri

From W3C Wiki
Revision as of 05:32, 14 December 2004 by (Talk)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

If you're trying to make a GoodUri, but for practical reasons can't achieve sufficient persistence guarantees, you may find redundancy can help.

Much like web content can be mirrored on many servers, you can have many servers hosting different URI names for the same thing, all stating that they are the same thing.

For example, we could identify a photograph as both

  http://site1.example.com/photo/43255

and

  http://site2.example.com/objX5FA691B

In RDF content using either name, we would try to add the triple

  <http://site1.example.com/photo/43255> owl:sameAs <http://site2.example.com/mirrors/objX5FA691B>

to make it clear they mean the same thing.

The observable behavior during dereference need not be the same, of course. One might be 404 while the other works. Or they might both work and redirect to completely different content; presumably the content for both would be relevant to the photograph in question. (Maybe one redirects directly to a JPEG, and the other to a page of photo metadata including a link to the JPEG.)

(Why "redirect"? Because a photograph is not a network information source. See SlashRedirection. Or we could use a HashURI, but then we couldn't us it in a browser (as a DualUseUri) and simply get the JPEG. We could get an HTML fragment with a <img ... > for the JPEG, though, so it's not a big difference. Hash or Slash, ... whatever.)