- From: Roy T. Fielding <fielding@apache.org>
- Date: Wed, 17 Jul 2002 15:21:12 -0700
- To: Tim Bray <tbray@textuality.com>
- Cc: www-tag <www-tag@w3.org>
I've read the RDF discussion and did not find anything there that hadn't been discussed before. Some people would like to limit the "http" namespace to the identification of documents, even though they are fully aware of hundreds of examples wherein an "http" URI is being used to identify things that are not even remotely documents. Logic states that the assertion "for all http URI x, x identifies document" is false if there exists one http URI for which it is false. Well, http://www.google.com/search?hl=en&ie=ISO-8859-1&q=send+gsm+message will lead you to multiple sites that use an http URI as a form submit destination and at no time is the user ever under the impression that the URI identifies the document that is returned in the 200 OK response. Likewise, we find at http://www.google.com/search?hl=en&lr=&ie=ISO-8859-1&q=web+robot+remote+control that there are dozens of independently deployed physical objects that can be identified and manipulated through "http" URI. Therefore, the assertion is false and this discussion is over. If the Semantic Web's tools cannot reason about these resources then those tools should be fixed. REST has absolutely no problem or ambiguity in modeling those identified resources as part of the real Web even though they do not act like an information space. This is not, in any way, a suggestion that all resources should be in the http scheme space. It merely proves that the arguments made about the range of http being limited are simply false and do not deserve any more of our time. Cheers, Roy T. Fielding, Chief Scientist, Day Software (roy.fielding@day.com) <http://www.day.com/> Chairman, The Apache Software Foundation (fielding@apache.org) <http://www.apache.org/>
Received on Wednesday, 17 July 2002 18:23:40 UTC