Re: issue 6401/6661: combined proposal

Attached is a modified version of the proposal. It includes the following 
edits:
1 - added ref to RFC3987 - per Ram's note
2 - fixed the poorly worded sentence in A2.1 - per Ram's note.
3 - added the A2.2.1 text Gil sent below.  Not being there seems cause 
some concern/confusion so I took the liberty of adding it back in - plus 
it seems pretty straight-forward.
4 - added "in Section A.2 and A.3" to section A - per Li's note.
5 - In the last sentence of A.2.1: s/EventDescription/EventDescriptions/ - 
just minor typo.

concerning some of the other points:
- Ram's comment about the cardinality - I can live with the current stuff. 
If it becomes a problem we can open a new issue later.
- Li's edit: "These <add>Events and</add>Notifications MAY be described 
via a Web Services Definition Language..."  is not quite correct.  WSDL 
describes the messages that flow over the wire.  In this case the messages 
are Notifications not Events. While its true that Events are usually 
someplace in the message, the WSDL may or may not describe them.  For 
example, in a wrapped notification the WSDL has nothing that describes the 
Events at all - its just an xs:any in a wrapper.  We can discuss this more 
during the next call.



thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com
The more I'm around some people, the more I like my dog.



Gilbert Pilz <gilbert.pilz@oracle.com> 
Sent by: public-ws-resource-access-request@w3.org
08/26/2009 04:34 PM

To
Ram Jeyaraman <Ram.Jeyaraman@microsoft.com>
cc
Doug Davis/Raleigh/IBM@IBMUS, "public-ws-resource-access@w3.org" 
<public-ws-resource-access@w3.org>
Subject
Re: issue 6401/6661: combined proposal






For the Binding for Unwrapped Notifications, I suggest using the text that 
I removed from a previous proposal:
The information about an Event Type contained in the wse:eventType element 
binds to a Unwrapped Notification for that type as follows:
The [action] property of the Notification has the value of the actionURI 
attribute of the wse:eventType element corresponding to the type of the 
Event being transmitted.
The [children] property of the Notification's Body element has a single 
child element. This child element is an instance of the Global Element 
Declaration referenced by the element attribute of the wse:eventType 
element corresponding to the type of the Event being transmitted.
As I stated on the last concall, I only removed the above because that is 
what we agreed to at the F2F.

As for MIME types, what is wrong with simply using "text/xml"?

- gp

On 8/26/2009 10:54 AM, Ram Jeyaraman wrote: 
This version generally looks good.
 
Some comments:
 
·       Add a reference to RFC 3987.
·       Appendix A.2.  Comment I2 about the cardinality of EventTypes 
needs to be addressed.
·       Appendix A.2.1. The sentence ?An Event Source there MUST NOT exist 
more than one EventDescription document? need to be correctly worded.
·       Appendix A.2.2.1 Binding for Unwrapped notifications is currently 
TBD. What is the plan for defining the binding? I prefer that the proposal 
either has the binding defined. If this needs more time, I suggest 
deleting the section and adding it later. This helps avoid confusion when 
we go into Last Call.
·       In the F2F discussions on this issue, we noted that we need to 
consider defining a MIME type for the Event Descriptions to be consistent 
with usage of other transmission types. I suggest this proposal include a 
MIME type for Event Descriptions.
 
Thanks.
 
From: public-ws-resource-access-request@w3.org [
mailto:public-ws-resource-access-request@w3.org] On Behalf Of Doug Davis
Sent: Monday, August 24, 2009 4:11 PM
To: Gilbert Pilz
Cc: public-ws-resource-access@w3.org
Subject: Re: issue 6401/6661: combined proposal
 

see if this version is any better.  We need to be careful because its not 
correct to say how many MetadataSections can appear, rather we need to 
talk about metadata documents.  For any one metadata document there could 
be at least 3 different MetadataSections returned (epr, uri, metadata). 


thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com
The more I'm around some people, the more I like my dog. 


Gilbert Pilz <gilbert.pilz@oracle.com> 
Sent by: public-ws-resource-access-request@w3.org 
08/24/2009 06:09 PM 


To
Doug Davis/Raleigh/IBM@IBMUS 
cc
"public-ws-resource-access@w3.org" <public-ws-resource-access@w3.org> 
Subject
Re: issue 6401/6661: combined proposal
 








I agree with the idea that we shouldn't repeat what MEX specifies, 
however, you have removed the parts where it states that the mex:Metadata 
for a single Event Source can contain at most one wse:EventDescriptions 
element and only one Notification WSDL per Notification 
Format/@Identifier. These are key constraints that need to be stated 
somewhere.

- gp

On 8/24/2009 1:38 PM, Doug Davis wrote: 

Overall I like the flow of this but I tried to remove some extra stuff 
that I think just adds verbosity w/o a good reason and will just lead to 
confusion ( like duplicating what MEX and WSDL already tell us).   
I also removed the unchanged portions of WS-Eventing - to keep it below 
the w3c mailing list size restriction.  :-) 



thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com
The more I'm around some people, the more I like my dog. 

public-ws-resource-access-request@w3.org wrote on 08/24/2009 02:23:32 PM:

> Attached (or inlined as the case may be) is draft 6 of a proposal 
> for issues 6401/6661. This proposal allows for the use of both the 
> EventDescriptions element and Format-specific Notification WSDLs. 
> There are still some open issues with this version of the proposal, 
> but these can be worked out by the WG. Note that, as per our 
> agreement at the last F2F, the section that describes the binding of
> wse:EventDescriptions to a Unwrapped Notification WSDL has been marked 
"TBD".
> 
> Thanks to Ram, Wu, and Li for their help and feedback. Thanks to 
> their input I think we've got something in which the combination of 
> EventDescriptions and Notification WSDLs offers some value beyond 
> merely serving as a political compromise.
> 
> - gp