- From: Paul Prescod <paul@prescod.net>
- Date: Mon, 12 Aug 2002 11:34:56 -0400
- To: Christopher B Ferris <chrisfer@us.ibm.com>, www-ws-arch@w3.org
Christopher B Ferris wrote: > >... > > My point was that you claimed that the OMG never felt the need for > "choreography". > I believe that the EDOC RFP belies your claim. ???? Let's define terms. Choreography is a way of describing the set of states that a multi-message stateful conversation can go through so that both participants know in advance what those states are. Now on to EDOC: "This RFP solicits proposals for a UML profile that supports the requirements for driving an object-oriented design of an enterprise computing system to an implementation in an enterprise distributed computing environment using an enterprise-class component model." * http://cgi.omg.org/docs/ad/99-03-10.txt I do not see how the EDOC RFP attempts to address the "choreography problem". It does not even mention the problem! Further, I claim that CORBA does not have a choreography problem. Good CORBA object design will enforce correct choreography in an explicit, statically-checkable, design-time-available manner. With good CORBA object design, a standard Java or C# static type checker can prevent you from calling messages in the wrong order, because choreography is enforced by the structure of the data and Java/C# type checkers enforce proper data structures. Similarly, XML and RDF schema languages can enforce proper data structures *explicitly*, *at design time* and at runtime so choreography is not necessary. -- "When I walk on the floor for the final execution, I'll wear a denim suit. I'll walk in there like Willie Nelson, John Wayne, Will Smith -- Men in Black -- James Brown. Maybe do a Michael Jackson moonwalk." Congressman James Traficant.
Received on Monday, 12 August 2002 11:37:27 UTC