W3C home > Mailing lists > Public > www-ws-desc@w3.org > July 2004

Re: Action Item 2004-07-01 Solution to 168/R114

From: Mark Baker <distobj@acm.org>
Date: Wed, 14 Jul 2004 16:29:02 -0400
To: Savas Parastatidis <Savas.Parastatidis@newcastle.ac.uk>
Cc: WS Description List <www-ws-desc@w3.org>
Message-ID: <20040714202901.GL28869@markbaker.ca>

Hey,

On Wed, Jul 14, 2004 at 06:15:04PM +0100, Savas Parastatidis wrote:
> > BTW, in case it helps explain the issue, I should clarify that by
> using
> > this style of message, the client can have no expectation that the
> quote
> > will be returned on the response to the "MSFT" message, because to
> have
> > such an expectation would be assuming a contract with "getStockQuote"
> > (*not* processMessage) semantics and would therefore have the same
> > non-self-descriptive problems as the other approach I'm critiquing.
> > 
> 
> I don't think we are disagreeing here. To expect that just because the
> operation is called "getStockQuote" you'll get the semantics which may
> implied by the human-readable name (without actually going into a
> specification and reading about the intentions of the service-providers)
> I think is incorrect. All you can expect is structural compliance to the
> service's description of the messages being exchanged.

Certainly.  I figured it was a given that there'd need to be some
human parseable specification about the interface semantics; for
example, the APIs_Reference.html that you can download from
http://www.google.com/api/

> Perhaps we should come up with a language called WSSL (WS Semantics
> Language) to accompany WSDL (WS Description Language).
> 
> > Note that Jim and Savas and I have talked about this extensively, but
> > they still disagree with me.
> > 
> 
> Yes! You are wrong and we are right :-))) (just joking of course)

Wow, I also agree that you are wrong and I am right!  What a
coincidence!  See, we're not so far apart.  8-) 8-)

But seriously, it sounds like we're closer than I remember us being.
It makes me wonder how you'd suggest queries be done with a
processMessage approach though.  But let's take that offline.

Mark.
-- 
Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca

  Seeking work on large scale application/data integration projects
  and/or the enabling infrastructure for same.
Received on Wednesday, 14 July 2004 16:28:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:32 GMT