2008/11/17 foodyzone <[EMAIL PROTECTED]>: > So, if that's what you mean by T-SOA, then I have to say that I can't agree > with your > original premise at all (though I must sound like a heretic saying so on > this email list!). > > By all accounts, T-SOA has been _wildly_ successful... with one caveat... > most people don't > call it "SOA". XML, REST, Web 2.0 / AJAX, SaaS, cloud computing - all are > applications of > service-oriented technical principles (T-SOA) and yet no one calls these > "SOA".
I agree in part, lots of people do call this stuff SOA. > > This likely explains why you hear people (whether business people or EAs) > saying "T-SOA > is unsuccessful". They are probably thinking only about WS-* (and I'd agree > on that > point). Nope, I've seen people already worried about REST and Web 2.0, and people who have SAP and Oracle are (IME) loving the WS-* stuff over the previous approaches. The key is that often they've been sold a "this will give you a magnitude improvement in performance" story from someone and it is failing to meet those objectives. In other words it has turned out that they are basically performing at a similar level just with new technologies. > I talk with lots of application teams that swear up and down they > aren't doing SOA, > and yet you actually look at what they are doing and it's riddled with > service-oriented > principles. T-SOA has become almost universal it seems. I agree. > > In constrast, let's look at the numbers for B-SOA: Gartner's latest survey > shows that > "SOA" adoption dramatically reduced in 2008. Taken in combinations with Anne > Thomas > Manes' analysis, that showed that the majority of people that say they are > doing "SOA" in > surveys aren't actually doing what we would consider "real" SOA (B-SOA), and > this paints a > pretty bleak picture for B-SOA. It could do, but one thing I wonder about these surveys is who gets asked. What I mean is that I'm seeing technology centric folks now jumping onto REST or something else and to be honest that isn't a surprise. What I'm also seeing is a new generation of CIOs and business folks (see the Economist Intelligence survey from 2006 or the Capgemini CIO Survey (I work there but its still good info) on this trend) who are viewing IT in a difference sense. I agree that most people aren't doing "real" SOA, I've just noticed that more people are now asking about it now that the technology is sort of "there" but not delivering the success that the vendors et al predicted. Steve > > - Dan Foody > > --- In [email protected], "Steve Jones" > <[EMAIL PROTECTED]> wrote: >> >> It depends what you mean. >> >> T-SOA for me is something pushed by vendors and those with vested >> interest in the vendors. It is the thing that says that XML is game >> changing and that all you need are a bunch of Web Services and BPEL >> and you are away. >> >> Clearly SOA has to have a sound technical architecture,but that >> certainly does not require the Web Services/BPEL/XML/REST >> implementation centric view. >> >> A sound Business driven SOA approach can have _exactly_ the same >> strong technical architecture as the previous set of solutions but >> have them now delivered more effectively as a result of the different >> structures, organisation, governance and commercials that delivers >> them. >> >> That certainly isn't BS, its what I'm seeing (and delivering). >> >> Steve >> >> >> 2008/11/15 Nick Gall <[EMAIL PROTECTED]>: >> > On Sat, Nov 15, 2008 at 5:00 PM, Ashley at Metamaxim >> > <[EMAIL PROTECTED]> wrote: >> >> I'm a bit confused too. >> >> >> >> Is it possible to imagine a T-SOA project that has a sound (realizable) >> >> business case? >> >> >> >> Or is any SOA project with a sound business case necessarily B-SOA? >> > >> > A sound SOA initiative will have BOTH a sound business case and a sound >> > technology architecture. Anything else is BS-SOA. >> > Does that make it clearer? >> > -- Nick >> > >> > >
