2009/6/9 Anne Thomas Manes <[email protected]>: > > > Hitoshi, > > My insurance company just changed its name from AIG to 21st Century. > My guess is that they lost many customers when AIG took the bail-out. > Sometimes changing names is the right thing to do. > > Please be aware that I have never advocated that we come up with a new > name to replace "SOA". My point is that we should never have attempted > to sell "SOA" to the business. It is a mistake to try to sell a vague, > abstract, IT architectural concept to business people.
This Anne is where we differ, I don't see SOA as a vague IT architectural concept but as a business model that is imposed onto the IT department. That is how I have "sold" it. With the simple questions Do you want IT to look like the business? Be managed like the business? Be costed in line with the business value? That is the transformation that a business SOA approach can deliver and the point to sell it is directly to the business as the transformation programme for IT. For the last couple of years I've been selling and consulting pretty much only to business people. What you are talking about sounds like the vendor driven JBOWS/ESB/SDP mindset which clearly shouldn't be sold to the business (or arguably to IT). > It's like > trying to sell Web 2.0 to business people. That doesn't mean that > architects should stop doing SOA, just stop trying to "sell" it. As > Nike would say, "Just do it." This I have a problem with as well. If you are creating business services and saying how they work then you need to do this with the business, you'll almost certainly have to call them "services", hell at work our "service" catalogue basically follows the SOA RM in terms of its hierarchy. > > Michael, no doubt, will assert that business people get service > orientation. And perhaps they do from a business perspective -- but > these folks don't necessarily understand how service orientation > applies to IT systems, and trying to explain it to them will most like > result in confusion and frustration. Their definition of "service" is > different from what architects/developers think of as a "service", > e.g., "billing" vs an application that prints bills. The business view is the right one, its IT's job to match it. > > If you have to sell something to business people, sell them the actual > "services" that will deliver value to them (i.e., their definition of > "service"). From an IT perspective, a service equates to a project. > "Doing SOA" means applying SO principles in the design and > implementation of that project. That is IT-SOA, not Business SOA. For me what you are saying is that don't do a VendorX approach of selling SOA technologies and implementation approaches to the business. I agree with this. What I don't agree with is that an SOA modelling approach to the business isn't what you should do and that a business SOA approach isn't something that can be sold to the business. Steve > > Anne > > On Mon, Jun 8, 2009 at 6:36 PM, htshozawa<[email protected]> wrote: >> >> >> That's exactly why we should stick with SOA instead of changing names. >> Changing names seems like we were defeated or worse, that we are just >> trying >> to sell the same of stuff under a different name. The point is there were >> some (many?) initiatives/projects that went sour but if we were a used car >> dealer, would we change the name because the dealership across the street >> is >> selling lots of lemons? >> >> I think it's just a matter of being able to explicitly show the benefits >> and >> show the differences between failed initiatives and our approaches like we >> always have been doing with proposals. IMO, sticking with the same name >> makes it easier to compare between those who really don't have any idea >> and >> those who do. It's easier to explain what's went wrong with the current >> initiative/project rather than not make users think that we are trying to >> pitch them the same thing under a different clothing. :-) >> >> H.Ozawa >> >> --- In [email protected], Anne Thomas Manes >> <atma...@...> wrote: >>> >>> Hitoshi, >>> >>> When I say SOA is dead, I mean that (in most organizations) business >>> people no longer believe the hype about SOA. The general attitude is >>> that SOA costs a lot and does not deliver value; therefore, funding >>> for SOA initiatives has dried up in most organizations. This is a >>> tragic development, because all organizations should be working to >>> optimize and improve their applications architecture. (Note, though, >>> that few so-called SOA initiatives were focused on architecture >>> improvement.) >>> >>> Given tight budgets and increased IT investment scrutiny, IT groups >>> should avoid putting forth proposals for "SOA" and instead focus on >>> developing proposals for concrete services with hard metrics that will >>> demonstrate quantifiable business value with rapid ROI. >>> >>> Anne >>> >> >> >
