Re: [soa-rm] David Linthicum Says: "ESB versus Fabric.Stop It!"

[Ken Laskey <klaskey@mitre.org>]

This isn't necessarily true because I can have a closed system that has  
a large, varied number of accessible services but no services are  
allowed from the outside world.  Many governance strategies assume this  
situation.  We get back again that if we have one service and one  
consumer and that service is currently sufficient for the needs of the  
consumer, do we have a SOA?

Ken


On May 25, 2005, at 8:48 PM, Rex Brooks wrote:

> Any self-contained system, is by definition, not SOA. It may have any  
> number of functional services within it, but as long as it is  
> self-contained it is not SOA. That's why I keep saying that the atomic  
> unit of SOA is a service AND a service consumer. I guess we have to  
> qualify it to the extent that the systems are essentially separate,  
> even if they may have endpoint interfaces that allow connection and  
> other connections are possible or even occur, but they are not  
> connected as an atomic unit until the service is invoked, or an  
> agreement to allow invocation has been reached.
>
> That is of course, just my opinion, even though I state it as if it  
> were fact.
>
> Ciao,
> Rex
>
------------------------------------------------------------------------ 
------------------
Ken Laskey
MITRE Corporation, M/S H305     phone:  703-983-7934
7515 Colshire Drive                        fax:        703-983-1379
McLean VA 22102-7508