Re: [wsbpel] Issue 110 - Rough draft for proposal to vote

["Yaron Y. Goland" <ygoland@bea.com>]

In/out are ambiguous because WSDL 1.1 uses them for both 
request/response and solicit/response. By using request/response we 
clarify exactly what we mean. I would also suggest that in common 
conversation programmers do not say 'in/out' they say request/response. 
I think aligning with common usage will make the spec clearer.

	Thanks,

		Yaron

Danny van der Rijn wrote:
> 
> 
> request-response has synchronous connotations that in/out typically has
> less of.  sure, i'm arguing small details here, but it's a small detail
> that your proposal changes.
> 
> as for alignment with WSDL, the messages themselves are called "in" and
> "out" in WSDL 1.0, while it is only their combination that is commonly
> (not normatively) referred to as request/response.  and request/response
> is not used (currently) in WSDL 2.0.  i realize that WSDL 2.0 may still
> change, so i'm not resting my argument on it, just offering it as a
> point of reference.
> 
> 
> Yaron Y. Goland wrote:
> 
>  > I'm unclear as to what connotations you refer to. The invoke activity
>  > can be legally used with two and only two MEPs in WSDL. The first is
>  > commonly referred to as a one-way. The second is commonly referred to
>  > as a request/response.
>  >
>  > The 'one-way' name doesn't provide any obvious inspiration for naming
>  > the variable that holds the outgoing message. So I instead took my
>  > inspiration from request/response which consists of a request and a
>  > response message.
>  >
>  > Therefore I do not believe that the change has any connotation besides
>  > aligning our language with WSDL which is the foundation of BPEL.
>  >
>  >         Yaron
>  >
>  > Danny van der Rijn wrote:
>  >
>  >>   maybe not semantics, but connotations, which i'd like to avoid
>  >> changing.
>  >>
>  >> Ugo Corda wrote:
>  >>
>  >>>> this ties our semantics too closely to HTTP-like semantics  
>  >>>
>  >>> I don't think that Yaron's new text changes in any way the current
>  >>> semantics.
>  >>>
>  >>>
>  >>> Ugo
>  >>>
>  >>> To unsubscribe from this mailing list (and be removed from the
>  >>> roster of the OASIS TC), go to
>  >>> 
> http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php. 
> 
>  >>>
>  >>>
>  >>>
>  >>> 
>  >>>
>  >> To unsubscribe from this mailing list (and be removed from the roster
>  >> of the OASIS TC), go to
>  >> 
> http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php. 
> 
>  >>
>  >
>  >
>  > To unsubscribe from this mailing list (and be removed from the roster
>  > of the OASIS TC), go to
>  > 
> http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php. 
> 
>  >
>  >
>  >
> 
> To unsubscribe from this mailing list (and be removed from the roster of 
> the OASIS TC), go to 
> http://www.oasis-open.org/apps/org/workgroup/wsbpel/members/leave_workgroup.php. 
> 
>