Re: Minutes, 1 May 2003 WS Description WG telcon

Arthur's statement regarding the importance of discovery in the Grid
applications we are looking at.
A single "type" of a service, as manifested by an explicit association
between a wsdl:interface and a wsdl:service would make searching within
registries (UDDI or otherwise) easier.

As it stands now, some of the service interfaces we have been building have
to return a list of interfaces supported.  Reducing this to a singleton, ie
THE interface implemented by the service, simplifies a couple of aspects of
the OGSI programming  model.

sgg

++++++++
Steve Graham
sggraham@us.ibm.com
(919)254-0615 (T/L 444)
Emerging Technologies
++++++++



                                                                                                                                       
                      "Arthur Ryman"                                                                                                   
                      <ryman@ca.ibm.com        To:       "Jonathan Marsh" <jmarsh@microsoft.com>                                       
                      >                        cc:       www-ws-desc@w3.org, www-ws-desc-request@w3.org                                
                      Sent by:                 Subject:  Re: Minutes, 1 May 2003 WS Description WG telcon                              
                      www-ws-desc-reque                                                                                                
                      st@w3.org                                                                                                        
                                                                                                                                       
                                                                                                                                       
                      05/05/2003 09:38                                                                                                 
                      AM                                                                                                               
                                                                                                                                       
                                                                                                                                       





Some amendments to the minutes:

1. I wasn't the one who said this:

... IBM\Arthur notes that some aspects of Schema are not well understood
and thus not implemented.

It's probably true but I don't recall who said that.

2. I mentioned a third benefit to the proposal for a service to implement a
single interface:

Searching for a service that implements a given interface, e.g. in UDDI,
becomes simpler if the service has an interface explicitly associated with
it. I believe this is important for GRID.


Arthur Ryman