Re: A(nother) Guide to Publishing Linked Data Without Redirects

On Fri, 2010-11-12 at 12:51 +0000, Graham Klyne wrote:
[ . . . ]
> The use of '#' indirection is one way to achieve this.
> 
> Another is Larry Masinter's tdb: URI scheme proposal 
> (http://tools.ietf.org/html/draft-masinter-dated-uri).

As a reminder, there is no need to define a new URI scheme to do this,
and defining new URI schemes is generally harmful.  See TimBL's writings
on "The Myth of Names and Addresses": 
http://www.w3.org/TR/webarch/#URI-scheme
and the W3C Architecture of the World Wide Web's recommendation:
http://www.w3.org/TR/webarch/#URI-scheme 
"A specification SHOULD reuse an existing URI scheme (rather than create
a new one) when it provides the desired properties of identifiers and
their relation to resources."

Instead of defining a new URI scheme, a similar effect can be obtained
-- but with the added benefit of dereferenceability -- by defining
specialized http URI prefixes, as described in "Converting New URI
Schemes or URN Sub-Schemes to HTTP:
http://dbooth.org/2006/urn2http/

One example of this is http://thing-described-by.org/ , which could do
nearly the same thing as the tdb scheme, but with the added benefit of
being layered on http, and hence directly deferenceable.  Instead of
writing:

  tdb:http://example/foo

one would write:

  http://t-d-b.org/?http://example/foo

syntactically the latter just has a longer prefix, but the added benefit
is that it *also* already works with today's http software.



-- 
David Booth, Ph.D.
Cleveland Clinic (contractor)
http://dbooth.org/

Opinions expressed herein are those of the author and do not necessarily
reflect those of Cleveland Clinic.

Received on Wednesday, 17 November 2010 21:31:23 UTC