W3C home > Mailing lists > Public > public-ws-semann@w3.org > January 2007

Re: [Fwd: Section on adding model references to inputs and outputs]

From: Ajith Ranabahu <ajith.ranabahu@gmail.com>
Date: Mon, 29 Jan 2007 10:23:18 -0500
Message-ID: <f43ea5790701290723h711d725fm30a9e4983439061a@mail.gmail.com>
Cc: "SAWSDL public list" <public-ws-semann@w3.org>

Hi,
> >
> > There is a problem with "this is equivalent to adding model references
> > to the relevant schema element". In particular, consider this scenario:
> > <types>
> >   <schema>
> >     <element name="ack"/>
> >   </schema>
> > </types>
> > <interface>
> >   <operation name="a">
> >     <output element="ack" modelReference="a_ack" />
> >   </operation>
> >   <operation name="b">
> >     <output element="ack" modelReference="b_ack" />
> >   </operation>
> > </interface>
> >
> > This would be equivalent to
> >   <element name="ack" modelReference="a_ack b_ack"/>
> > which is not what you want, if I understand it correctly.

well the current model allows that but my feeling is that once the
model reference is taken into account it becomes less obvious that the
concept 'a_ack'  is relevant in the scope of the operation 'a' and
'b_ack' becomes active in the scope of operation 'b'. When both of
thes e annotations are placed in the element itself it applies to the
element universally  - not in a specific context.

> > Apart from this, your proposal sounds reasonable, yet I would like to
> > see more details - in particular, how exactly would your modelReferences
> > on the two ack outputs differ? Can you please give us the WSDL annotated
> > with any SAWSDL annotations that you would use there, so it's clearer
> > what exactly you want to do with them?
> >
 yes - I will have a detailed example and post it later today.

-- 
Ajith Ranabahu
Received on Monday, 29 January 2007 15:31:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 17 April 2012 12:14:27 GMT