Comments on: An alternative to canonical URIs http://www.xaprb.com/blog/2006/06/02/an-alternative-to-canonical-uris/ Stay curious! Fri, 08 Aug 2008 18:35:34 +0000 http://wordpress.org/?v=2.2.2 By: Xaprb http://www.xaprb.com/blog/2006/06/02/an-alternative-to-canonical-uris/#comment-825 Xaprb Sat, 03 Jun 2006 13:32:32 +0000 http://www.xaprb.com/blog/2006/06/02/an-alternative-to-canonical-uris/#comment-825 <p>Well, parameters are unavoidably ordered by their position in the query string, but most software that uses them treats them as un-ordered. The <abbr title="Request For Comments">RFC</abbr>s I can find don't specify anything about ordering. I think that is left up to the program that processes the request. I'm looking at <a href="http://tools.ietf.org/html/3986" rel="nofollow">RFC 3986</a>, <a href="http://tools.ietf.org/html/1738" rel="nofollow">RFC 1738</a>, and <a href="http://tools.ietf.org/html/1630" rel="nofollow">RFC 1630</a>. Some of those have been updated or obsoleted by other RFCs, but I don't see where any of them says anything other than "The query component is a string of information to be interpreted by the resource" (from <a href="http://tools.ietf.org/html/2396" rel="nofollow">RFC 2396</a>).</p> Well, parameters are unavoidably ordered by their position in the query string, but most software that uses them treats them as un-ordered. The RFCs I can find don’t specify anything about ordering. I think that is left up to the program that processes the request. I’m looking at RFC 3986, RFC 1738, and RFC 1630. Some of those have been updated or obsoleted by other RFCs, but I don’t see where any of them says anything other than “The query component is a string of information to be interpreted by the resource” (from RFC 2396).

]]> By: Tim McCormack http://www.xaprb.com/blog/2006/06/02/an-alternative-to-canonical-uris/#comment-823 Tim McCormack Sat, 03 Jun 2006 04:14:02 +0000 http://www.xaprb.com/blog/2006/06/02/an-alternative-to-canonical-uris/#comment-823 <p>This feels like a much better solution than the canonicalization approach (though I suppose it doesn't deal with www. versus bare domains.)</p> <p>Repeated parameters could be handled with character code order. Isn't the parameter list independent of order?</p> This feels like a much better solution than the canonicalization approach (though I suppose it doesn’t deal with www. versus bare domains.)

Repeated parameters could be handled with character code order. Isn’t the parameter list independent of order?

]]>