FW: SOA Reference Model

["Duane Nickull" <dnickull@adobe.com>]

Some praise of our work and thoughts for the RA...courtesy of Jos.

*******************************
Adobe Systems, Inc. - http://www.adobe.com
Vice Chair - UN/CEFACT  http://www.uncefact.org/
Chair - OASIS SOA Reference Model Technical Committee
Personal Blog - http://technoracle.blogspot.com/
******************************* 

> -----Original Message-----
> From: Duane Nickull [mailto:dnickull@adobe.com] 
> Sent: woensdag 1 februari 2006 16:25
> To: Vrancken, Jos
> Subject: RE: SOA Reference Model
> 
> 
> Hello Jos:
> 
> First - thank you for taking the time to write in.  Now I am 
> curious to read your presentation ;-)
> 
> The SOA RM is purely abstract, hence there is no concrete 
> aspects of a service to use as metrics such as 
> generality/specialized.  Any mention of such would be merely 
> incidental.  
> 
> However - we do plan to do this in greater detail in the SOA 
> Reference Architecture.  Your layered diagrams would be most 
> useful to that sub group and I would ask if I may forward 
> your email to that group.  I am sure they would appreciate them.
> 
> Duane
> 
> *******************************
> Adobe Systems, Inc. - http://www.adobe.com
> Vice Chair - UN/CEFACT  http://www.uncefact.org/
> Chair - OASIS SOA Reference Model Technical Committee
> Personal Blog - http://technoracle.blogspot.com/
> ******************************* 
> 
> 
> -----Original Message-----
> From: Vrancken, Jos [mailto:j.vrancken@tbm.tudelft.nl] 
> Sent: Wednesday, February 01, 2006 6:30 AM
> To: Duane Nickull
> Subject: SOA Reference Model
> 
> Hello Duane,
> I just read your presentation on the SOA Reference Model.
> Very interesting indeed!
> It has several points in accordance with a presentation that
> I gave recently on application integration using semantic
> web services, and that is 
> usually a very positive sign, as we had no prior contacts about this. 
> Points of accordance were
> - importance of layer diagrams
> - crisscross relationships between adjacent layers in layer diagrams
> - the two aspects behavior and information of a service
> - the importance of architecural models
> 
> 
> May I make a proposal for the core definition of SOA.
> That is about the service notion. 
> 
> The point is: useful things are both very general and very 
> specific. The service notion as defined in your presentation 
> is sufficiently general but it still lacks, IMHO, specificity. 
> We can make it more specific without falling in the
> trap of specific technologies. 
> The whole subject is about implementing services in software. 
> My proposal for the definition of a service is: a service is 
> the capability of a software component with a layered 
> interface. The layeredness applies to both aspects of 
> the interface: the information and the behavior.
> Good interfaces are essential for effective modularization
> and decoupling of components. So is layeredness. And when
> the interfaces themselves are layered, that is probably the best 
> one can achieve in terms of modularity, independence and 
> reduced change propagation.
> 
> In the case of Web Services the two layer diagrams,
> for information and for behavior may look as follows:
> 
> Information:
> 
> _______________
> specific format
> _______________
> XML-Schema
> _______________
> XML
> _______________
> Unicode
> _______________
> Bytes
> _______________
> 
> 
> 
> Behavior:
> 
> ________________________________
> specific interaction
> ________________________________
> composite interaction
> ________________________________
> synch. and asynch. communication 
> ________________________________ asynch interaction 
> ________________________________
> 
> 
> 
> Could this idea be of some value for the SOA Reference Model? 
> Then I would like to participate in some way in your TC.
> 
> 
> Kindest regards / Met vriendelijke groet,
> 
> Jos Vrancken
>  
> Sectie ICT, TBM, TU-Delft
> Jaffalaan 5
> 2628 BX Delft
> P.O. Box 5015
> 2600 GA Delft
> telephone 06-51402799 / 015-2781566
> room B2.310
> 
>