[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Web Services/SOA (was RE: XML 2004 weblog items?

• To: 'Karl Waclawek' <karl@w...>, xml-dev@l...
• Subject: RE: Web Services/SOA (was RE: XML 2004 weblog items?)
• From: "Bullard, Claude L (Len)" <len.bullard@i...>
• Date: Tue, 30 Nov 2004 13:14:06 -0600

I can see that as philosophy, but practically, it 
doesn't make a difference other than performance. 
In work we are doing, we are using web services 
for things such as report methods.   The fact of 
locale makes no difference.  It's just a report. 
Where locale *might* be factored in is roles and 
permissions.

To me, the SOA as philosophical architecture is 
an organizational principle, that is, as an enterprise 
engineer, I intuitively understand the concepts of 
breaking tasks into services thus enabling a contract 
model of command, control and communication, and yes, 
chunkier tasks.  

As a programmer, it looks like an encapsulated 
method that advertises an interface.

The former is why I tend to think in terms of events. 
Then notions such as correlation engines make more 
sense (see autonomic computing) because I can have 
event ontologies for reasoning about the event type 
and assembling a service to be performed from a 
repetoire of services enterprises of this.type offer.

IMHO, the service model fell out of the enterprise 
modeling domain.  It is how businessHeads see their 
businesses, as contracts for products, goods, and services.

len


From: Karl Waclawek [mailto:karl@w...]

> Are services responses to events?
> 
> That is likely at least one level higher 
> in the organizational architecture if call 
> and response is the lowest level of description, 
> but if we are to speak of a 'services oriented 
> architecture' that is meaningful beyond the 
> most primitive descriptions, it can be useful 
> to think in terms of event types over  
> messages.  Otherwise, a service and a method 
> are indistinguishable.

I have a hard time seeing something new and unique in SOA
worthy of inventing a new acronym.

Maybe the difference is in the fact that services
have "coarser granularity", they do more per request,
they are "chunkier". Which may mean that what they
provide is just that much closer to what a non-programmer
is able to describe, that they deserve their own terminology.

Another "gradual" difference from - let's say, distributed
objects - could be that in SOA there is a heightened awareness
of remote interactions being fundamentally different from local
ones - with SOA specializing in the remote type.

• Follow-Ups:
  • Re: Web Services/SOA (was RE: XML 2004 weblog items?)
    • From: "Karl Waclawek" <karl@w...>

• Prev by Date: RE: Web Services/SOA (was RE: XML 2004 weblog items?)
• Next by Date: RE: Web Services/SOA (was RE: XML 2004 weblog items?)
• Previous by thread: RE: Web Services/SOA (was RE: XML 2004 weblog items?)
• Next by thread: Re: Web Services/SOA (was RE: XML 2004 weblog items?)
• Index(es):
  • Date
  • Thread