W3C home > Mailing lists > Public > www-webont-wg@w3.org > November 2002

Formally specifying HTTP GET operations

From: Jonathan Borden <jonathan@openhealth.org>
Date: Wed, 27 Nov 2002 08:12:37 -0500
Message-ID: <019801c29616$a845f880$7c674544@ne.mediaone.net>
To: <www-webont-wg@w3.org>

There are several points in which we desire to formally specify an HTTP GET
operation. When we wish to assert the 'contents' of a URI, we wish to assert
the contents of an RDF document obtained when dereferencing the URI. This
exposes the need to formally define the HTTP GET operation itself.

An HTTP GET is composed of two datastructures: an HTTP GET request and an
HTTP GET response. We can characterize the entailment of the contents of a
URI as defining an entailment from an HTTP GET request to an HTTP GET
response. Representing the HTTP request and response messages as RDF should
ease the 'fit' of this type of entailment into RDF entailments.

In 1998 I first described an XML representation of MIME -- the datastructure
for SMTP and HTTP messages. http://www.openhealth.org/xmtp/ This
representation has been tested by operating software which performs this
conversion since 1998. More recently I updated the format to RDF. This RDF
representation of MIME messages may be used to characterize an HTTP GET
operation this: http://www.openhealth.org/xmtp/HTTP/

Formally defining an HTTP GET operation might be out of the scope of OWL,
but hopefully this description will demonstrate some of the issues in
defining OWL entailments that cross HTTP GET operations.

Jonathan

http://www.jonathanborden-md.com
http://www.erieneurosurgery.com
http://www.openhealth.org
Received on Wednesday, 27 November 2002 08:32:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:55 GMT