I believe TAFKAS is more than a pattern and approach. Following the architectural process, the resulting architecture is a physical thing can be judged as to its degree of service orientation. I understand this is an idea still without consensus on this forum. Oh, TAFKAS of course is another SLA (six letter acronym) meaning The Architecture Formerly Known as SOA).
So, yes, with an eye on the resulting architecture, it absolutely includes technology peices IMO. I personally feel that every dimension of SOA, errr TAFKAS, has been oversold. Not just the tools and technologies by vendors, but the processes as well by consultants. To me, service orientation is an evolutionary (not revolutionary) concept of applying the principles of interchangeable parts. It's the next stepping stone beyond OO and component-based development of systems. Alignment of "parts" to business concepts is an important bonus, but its still just interchangeable parts. It's hard to sell a simple evolutionary step, thus the hype on all fronts. Other idustries have gone through this evolution, and I believe found the same limitations: interchangeable parts are easy in the small, but exponentially more difficult in the large. Tires and wiper blades are easily replaced, but engines are not. The challenge remains for us how to solve the problem in the large, regardless of what we call the concept. -Kirstan --- In [email protected], David Chappell <david.chapp...@...> wrote: > > I think you and others on this list have beat that one to death. The architecture until recently known as SOA is an architectural pattern, a journey, an approach to architecture, an approach to orgranization, etc that is about how applications are being built going forward AND about how IT aligns with the business. > > However I have to disagree with your continued implications that SOA is unrelated to tools and technologies. Anyone who designs a SOA without tools and technologies in mind is being pretty shortsighted. > Dave > > > _____ > > From: Michael Poulin [mailto:m3pou...@...] > Sent: Wednesday, January 07, 2009 1:25 PM > To: [email protected] > Subject: Re: [service-orientated-architecture] Re: SOA is Dead > > > > > +100 to "The problem, IMO, isn't that the concepts and principles are flawed. It's that "SOA" in the minds of many is synonymous with specific tools and technologies. " > > This is because people tried to sell HOW instead of WHAT and WHY > > - Michael > > > _____ > > From: Rob Eamon <rea...@cableone.net> > To: service-orientated-architect...@yahoogroups.com > Sent: Wednesday, January 7, 2009 4:35:05 PM > Subject: [service-orientated-architecture] Re: SOA is Dead > > > > The problem, IMO, isn't that the concepts and principles are flawed. > It's that "SOA" in the minds of many is synonymous with specific > tools and technologies. They deploy an ESB and expect all kinds of > benefits which have nothing to do with ESBs. The "EAI" term suffered > the same perception problem--EAI still does for folks like Steve > Jones, even though many of the principles behind EAI are the same as > for SOA. :-) > > "It always gets back to discussions of definitions, and approaches." > > That's what architecture is about, no? > > SOA is definitely not a clearly defined path. It is an architectural > approach with many, many possible paths. > > IT should *never* try to "sell" anything at all to the business. > > -Rob > > --- In HYPERLINK "mailto:service-orientated-architecture% 40yahoogroups.com" \nservice-orientated- architecture@ yahoogroups. com, "Nibeck, > Mike" <mike.nibeck@ ...> wrote: > > > > "My point is that IT groups should no long attempt to > > sell "SOA" to the business. "SOA" is now a bad word.' > > > > So by changing what we call it, it will somehow become relevant and > > useful? Couldn't we be seeing a core, fundamental problem with SOA > > as a viable solution. Regardless of why or where it is breaking > > down, I think it's safe to say that it is frequently breaking > > down. Not all the time, as members of this can attest to, but if > > you look at the ratio of good core content vs.. semantic > > discussions of this list over the last 4+ moths, very little actual > > details are emerging. It always gets back to discussions of > > definitions, and approaches. > > > > The move from client-server to web worked not only because it was > > "better", but because it was a clearly defined, different approach > > to solution development. SOA (or whatever you call it) is too > > similar to past efforts, and isn't a clearly defined path. > > > > Just MHO. > > > > _mike >
