W3C home > Mailing lists > Public > www-archive@w3.org > June 2003

Re: targetResource wording

From: David Booth <dbooth@w3.org>
Date: Fri, 20 Jun 2003 12:51:22 -0400
Message-Id: <>
To: Mark Baker <distobj@acm.org>
Cc: www-archive@w3.org

Hi Mark,

In response to 

I thought I would add further precision to my explanation, but I don't know 
how many others will be interested in reading it, so I'm just sending it to 
you individually and copying the public archive, in case the question 
arises again.

If WSDL description d1 states "targetResource='u'", where u is some URI, 
then it is making the following assertions:

         There exists a service s1 described by d1.
         There exists a resource r identified by u.
         There exists a relationship e1 between s1 and r.

However, the WSDL 1.2 spec gives you no more information about what e1 or r 
are.  If WSDL description d2 also states "targetResource='u'", then it is 
adding the following assertions:

         There exists a service s2 described by d2.
         There exists a relationship e2 between s1 and r.

 From which you can conclude:

         There exists a relationship e between s1 and s2, such that
         ( e1(s1, r) and e2(s2, r) ) => e(s1, s2).

Again, the WSDL 1.2 spec gives you no further information about what that 
relationship e is.  You need to know the (application-defined) semantics of 
d1 and d2 to find out.  In the printer example, the semantics of d1 and d2 
might indicate something like:

         r = HP 3200 LaserJet printer with serial number 1234556234
         e1 = is-the-printing-service-for
         e2 = is-the-management-service-for
         e = is-the-management-service-corresponding-to-printing-service

Does that make sense?

David Booth
W3C Fellow / Hewlett-Packard
Telephone: +1.617.253.1273
Received on Friday, 20 June 2003 12:54:05 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:42:25 UTC