Re: i0001: EPRs as identifiers - alternative proposal

David Booth wrote:

>Tom,
>
>The reason is that Reference Properties are intended to be used to
>address things that -- for all intents and purposes -- represent
>different Web resources.  (I.e., they have different WSDL, metadata,
>policy, etc.)  Reference Parameters *could* be used that way -- just as
>cookies *could* be used that way -- but they are not *intended* to be
>used that way, nor are they most often used that way.  
>
>
>  
>
If you place a ref parm "<username joe/> in a soap header block where 
the message body has operation with semantic (what is your
birthdate) will you get a differenent answer from another exchange with 
everything else in the message the same but with
the ref parm value <username joesFather />?

Tom Rutt

>On Wed, 2004-12-08 at 21:16, Tom Rutt wrote:
>  
>
>>David Booth wrote:
>>
>>    
>>
>>>The question is: What guidance are we giving to the world?
>>>
>>>As you point out, we cannot stop people from doing things in ways that
>>>are anti-Web, nor should we.  For whatever reasons, sometimes someone
>>>may *need* to address Web resources using something other than a URIs.
>>>
>>>But what *guidance* are we giving?  If Reference Properties are kept in
>>>the spec, we would be *endorsing* a practice that is harmful to the Web
>>>as a whole.
>>>
>>>On the other hand, if we drop Reference Properties, people who really
>>>need to address Web resources using non-URIs could still (ab)use
>>>Reference *Parameters* to accomplish the same result, just as they
>>>sometimes use cookies to do so today.  
>>>
>>>      
>>>
>>I do not understand why Reference Parameters would be acceptable but 
>>Reference Properties are not?
>>
>>If either is present, they are used to identify the instance of what is 
>>being communicated to?
>>
>>Tom Rutt
>>Fujitsu
>>
>>    
>>
>>>Thus, they would not be prevented
>>>      
>>>
>>>from doing what they may need to do, but they would be encouraged to do
>>    
>>
>>>things the Web way.
>>>      
>>>
>
>  
>

-- 
----------------------------------------------------
Tom Rutt	email: tom@coastin.com; trutt@us.fujitsu.com
Tel: +1 732 801 5744          Fax: +1 732 774 5133

Received on Thursday, 9 December 2004 23:01:18 UTC