Comments on: The #referent Convention http://www.marcdegraauw.com/2007/02/20/the-referent-convention/ Some thoughts on meaning, the web and everything Wed, 20 Aug 2008 18:32:46 +0000 http://wordpress.org/?v=2.2.2 By: marc http://www.marcdegraauw.com/2007/02/20/the-referent-convention/#comment-403 marc Mon, 05 Mar 2007 14:29:30 +0000 http://www.marcdegraauw.com/2007/02/20/the-referent-convention/#comment-403 Lars, "... this solution ... violates the principle that URIs should be opaque." There is no principle that URIs should be opaque. As a URI-publisher I am free to define whatever non-opaque URI scheme I want, so I can say http://www.marcdegraauw.com/bestbeerbars?city=Amsterdam and tell users to fill in whatever city they want. See http://www.w3.org/2001/tag/doc/metaDataInURI-31#erroneous where it says: "Web software MUST NOT depend on the correctness of metadata inferred from a URI, except when the encoding of such metadata is documented by applicable standards and specifications. Such standards and specifications include ... documentation provided by the URI assignment authority." Lars,

“… this solution … violates the principle that URIs should be opaque.”

There is no principle that URIs should be opaque. As a URI-publisher I am free to define whatever non-opaque URI scheme I want, so I can say http://www.marcdegraauw.com/bestbeerbars?city=Amsterdam
and tell users to fill in whatever city they want.

See http://www.w3.org/2001/tag/doc/metaDataInURI-31#erroneous
where it says: “Web software MUST NOT depend on the correctness of metadata inferred from a URI, except when the encoding of such metadata is documented by applicable standards and specifications. Such standards and specifications include … documentation provided by the URI assignment authority.”

]]> By: Lars Marius Garshol http://www.marcdegraauw.com/2007/02/20/the-referent-convention/#comment-346 Lars Marius Garshol Fri, 23 Feb 2007 11:37:25 +0000 http://www.marcdegraauw.com/2007/02/20/the-referent-convention/#comment-346 The main problem with this solution is that it violates the principle that URIs should be opaque. That is, you have to look at the syntactic form of the URI to work out what it refers to. It's better than httpRange-14, though. But then, everything is better than httpRange-14, including no solution, because when we had no solution, at least we agreed we had no solution... The main problem with this solution is that it violates the principle that URIs should be opaque. That is, you have to look at the syntactic form of the URI to work out what it refers to.

It’s better than httpRange-14, though. But then, everything is better than httpRange-14, including no solution, because when we had no solution, at least we agreed we had no solution…

]]>