W3C home > Mailing lists > Public > www-ws-desc@w3.org > June 2002

Re: proposal for resolving service type issues

From: <sggraham@us.ibm.com>
Date: Tue, 11 Jun 2002 05:05:02 -0400
To: "Jean-Jacques Moreau" <moreau@crf.canon.fr>
Cc: Sanjiva Weerawarana <sanjiva@watson.ibm.com>, "WS-Desc WG (Public)" <www-ws-desc@w3.org>, www-ws-desc-request@w3.org
Message-ID: <OF9405A639.1515790C-ON85256BD5.0031DAF0@us.ibm.com>
One way around this is to make the implements element a child of the

Steve Graham
(919)254-0615 (T/L 444)
Emerging Technologies

                      Moreau"                  To:       Sanjiva Weerawarana <sanjiva@watson.ibm.com>                                  
                      <moreau@crf.             cc:       "WS-Desc WG (Public)" <www-ws-desc@w3.org>                                    
                      canon.fr>                Subject:  Re: proposal for resolving service type issues                                
                      Sent by: www-ws-                                                                                                 
                      06/10/2002 01:30                                                                                                 

Sanjiva, I like your proposal. In particular, I think there is a
lot of merit in bringing the notion of an abstract service into
the foreground. Please find my comments below.


> We require that all services defined in a single document be of
a single
> service type.

This is fine...

> That type is indicated by inserting the following required
>      <implements serviceType="qname"/>

... however I don't think we can require the wsdl:implement
element to be always present. For example, the WSDL file may only
contain an abstract service declaration, which is refined and
implemented in a second WSDL file. I think wsld:implement should
be optional unless there is a concrete service definition (i.e.
binding), in which case it should be mandatory.

> two portTypes are said to be equivalent iff they have the same
> name.

Hmmm... somebody could get it wrong and you could end up with two
portType with the same qname but different children EIIs. It's
probably not our business, though, and more like a programmer's

(image/gif attachment: graycol.gif)

(image/gif attachment: ecblank.gif)

Received on Tuesday, 11 June 2002 12:18:53 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:06:23 UTC