W3C home > Mailing lists > Public > www-ws-desc@w3.org > April 2004

RE: Issue 64 resolution: HTTP operations as WSDL operation names

From: David Orchard <dorchard@bea.com>
Date: Wed, 21 Apr 2004 20:25:05 -0700
Message-ID: <32D5845A745BFB429CBDBADA57CD41AF06F1ECDC@ussjex01.amer.bea.com>
To: "Roberto Chinnici" <Roberto.Chinnici@Sun.COM>
Cc: <www-ws-desc@w3.org>

These are separate points so I'll do separate responses for each.

The reason why ATOM:GET sucks is that means that every single vocabulary that wants to use GET semantics has to define GET for themselves.  It's as valuable for WSDL to define what "GET" means as it is valuable to define, say, wsdl:binding.  So that tools can have a common understanding of the what the defined names mean and their constraints.   

And why should Atom define Atom:GET?  How in any way are the semantics of HTTP GET specific to Atom?  The semantics of HTTP GET, and the ability for Web services to use HTTP GET at the interface level is usable across just about any Web service, and thus belongs in WSDL.

Cheers,
Dave

> -----Original Message-----
> From: Roberto Chinnici [mailto:Roberto.Chinnici@Sun.COM]
> Sent: Wednesday, April 21, 2004 6:03 PM
> To: David Orchard
> Cc: www-ws-desc@w3.org
> Subject: Re: Issue 64 resolution: HTTP operations as WSDL operation
> names
> 
> 
> David Orchard wrote:
> > 
> > 
> >>-----Original Message-----
> >>From: Roberto Chinnici [mailto:Roberto.Chinnici@Sun.COM]
> >>Sent: Wednesday, April 21, 2004 5:13 PM
> >>To: David Orchard
> >>Cc: www-ws-desc@w3.org
> >>Subject: Re: Issue 64 resolution: HTTP operations as WSDL operation
> >>names
> >>
> >>
> >>I'm not sure what concrete problem is addressed by this 
> proposal (and
> >>yes, I've read issue 64).
> >>
> >>In what sense is the operation called 
> >>QName("http://example.org/HTTP", 
> >>GET) *the* HTTP GET "operation" (= method, I presume)? Certainly I'm
> >>free to bind it in a completely HTTP-free way, say to 
> SOAP/SMTP. What
> >>does it mean in that case? And what if I bound it to an 
> http:operation
> >>with method="PUT"? Or is the REST:GET operation what people 
> are really
> >>asking for?
> > 
> > 
> > The operation is the GET operation in the same way that a 
> spec like Atom describes it. 
> 
> Atom:GET then? Sure we don't prevent Atom from defining one. What's so
> valuable in defining it in WSDL 2.0?
> 
> 
Received on Wednesday, 21 April 2004 23:25:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:30 GMT