OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

soa-rm message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [soa-rm] Service Consumer in RM or not?

Chiusano Joseph wrote:

> Duane,
> This is an idea that I see you have been pushing very hard almost from 
> the start of our TC, yet I believe some of us are perplexed as to why 
> introduction of a service consumer into an RM is against the notion of 
> RM. Can you please clarify for us?

With pleasure.  The basic tenet of a Reference Models are that they are 
abstract.  If we show both a service provider and a service consumer, 
then we have "infrastructure" which is now architecture rather than a 
model.  If we have an infrastructure, it is concrete rather than abstract.

This is a commonly used convention for most software architecture.  Once 
again, I will point at the OSI seven layer stack. There is not a 
"sender" and a "receiver" even though it is a communication stack 
model.  If it did, it would no longer be a reference model.

We (this TC) do need to demonstrate where and how an SC fits in with the 
notion of the RM.  I agree with this given the confusion.  The plan I 
would advocate is that we finish our Reference Model without the SC in 
the core model, then demonstrate in Appendix "B" how one may use the 
reference model to develop architecture (very simple example).  In this 
example, we can demonstrate that the SC is present and show how the data 
model becomes the "message", how the contract is formed etc.

We also have the option of creating a Reference Architecture, which can 
be built using WS-* standards.    This is probably the work that I think 
many will feel is very valuable for their specific purposes.  IMO - this 
is a better way to deliver our message since it will be in alignment 
with conventions used by many software architects.

The reason you may have perceived that I have pushed hard on this is 
that I see a dangerous turn of events if this TC single handedly tries 
to re-define the notion of Reference Models to be concrete rather than 
abstract.  IMO - our work would be largely ignored, or worse, it could 
becomes an example of how not to build RM's.

I hope this clarifies this.



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]