W3C home > Mailing lists > Public > public-ws-addressing@w3.org > November 2004

RE: Mandator wsa:Action (was Re: WS-Addr issues)

From: Francisco Curbera <curbera@us.ibm.com>
Date: Sat, 6 Nov 2004 22:10:58 -0500
To: "Savas Parastatidis" <Savas.Parastatidis@newcastle.ac.uk>
Cc: public-ws-addressing@w3.org, public-ws-addressing-request@w3.org
Message-ID: <OF8F32D175.800302AB-ON85256F45.000F1AC2-85256F45.00117C0E@us.ibm.com>


The semantics of WSA:Action are already defined in the current spec. The
acceptable values are defined by each service contract.l We have several
specifications already using it and several vendors interoperating on that
basis. I don't get exactly what is undefined about that.


                      "Savas Parastatidis"                                                                                                      
                      <Savas.Parastatidis@newca        To:       <public-ws-addressing@w3.org>                                                  
                      stle.ac.uk>                      cc:                                                                                      
                      Sent by:                         Subject:  RE: Mandator wsa:Action (was Re: WS-Addr issues)                               
                      11/05/2004 02:15 PM                                                                                                       

Hi Dave,

> I agree that there are cases where in a mandatory action will result
> garbage being filled in.  Same way as mandatory zip codes get filled
> with a "11111" by me as a resident of Canada.
> I can now imagine some viewers thoughts "Aha! See, you just proved my
> point about mandatory fields resulting in garbage.  How could you be
> foolish?!"
> There is a trade-off to be made.  Any mandatory fields may result in
> garbage.  Any optional fields may result in no data or even garbage.
> The trade-off is around certainty of the field with possibility of
> garbage versus lack of certainty of the field and still possibility of
> garbage.


I think this is a very good analysis! Thanks.

However, I believe that your analogy between wsa:action and the
mandatory ZIP code field on a web page has a very small flaw. In the
case of a ZIP code the semantics are clear. It's a zip code. If one
chooses to input 11111, that's their choice. The service provider will
check the validity of the data since the semantics of the field and its
acceptable values are well known.

What are the semantics of a wsa:action information header when its value
is not garbage? How could one determine by just looking at the
wsa:action header whether its value is garbage or not?

Of course one could define that the semantics of wsa:action are protocol
specific and the accepted values (not format of those values, since we
know it's going to be a URI) are not known. It's an open set. So what
does it mean to have urn:process:message vs
http://bla.bla/call/this/method? Of course we need separate
specifications to define the semantics of the _values_ inside a
was:action information header but what about wsa:action itself?

Savas Parastatidis
Received on Sunday, 7 November 2004 03:40:16 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:04:07 UTC