I think the industry is slowly learning, but the net effect is that by the time most of the "old timers" from the previous wave have learned the lessons there's all of the "new kids" with shinier toys joining the ranks that are supposed to make all of those problems which were starting to get under control not even be an issue anymore.
I was actually thinking about this in the shower today, but with the "Web/Enterprise/Government 2.0" hype instead of SOA or EAI, and I think a lot of the same reasoning applies. The focus is always about XYZ technology can help you solve ABC business problem. I think that's backwards. The same sort of business problems have been around forever. Sure, the cycle times are getting shorter and shorter, but the core problems are the same, and for some reason, we're still trying to solve them. Why? They're hard problems. They're *business* problems, and they're *people* problems that have very little to do with technology. However, the latest toys with the polished chrome and the fresh paint are enough to distract everyone long enough that the lessons either don't sink in, or they sink in too well. It's a subtle difference, but I think we (on the IT side) should spend more time talking and thinking about the business problems and which of these 10,000,000 technology options we have to solve them will give the right solution in the context of a given business environment. Bear in mind that I'm not talking about "business problems" in the sense of this is what a business user has said they wanted a system to do. Writing this now makes me think of the bit in Christensen about the vendors who listened to their customers were the ones that actually lost out. Everything you hear, and all of the "modern" development methodologies, have been talking about getting IT closer to the business, more agility and greater delivery speed. However, what if that's actually all wrong? I've a diagram I came up with for a training course I'm supposed to give in the next couple of months that I think finally makes it all fit together for me. I've been thinking about blogging it, so maybe this'll motivate me a bit more. The key thing I think I've figured out is that if you really want to deliver long-term business value with technology, you've got to align those technology investments with the vision--not the marketing gimmick, but the *real* vision for the company. I think this really is the only way you can deliver core capabilities rather than technical solutions. The business strategy is actually going to wander all over the place trying to best figure out how to capitalize on the business opportunities which are compatible with the vision. Most business people recognize this is simply a fact of business. So, if you create a technology strategy that's aligned with the business strategy de jour, you're only going to be chasing the rainbows and never really be able to focus on building those core capabilities. You might get lucky from time to time, but you're still chasing a moving target--like trying to chase a jackrabbit. It takes real leadership and a clear, easy to articulate business vision, but I think that it is possible to identify a way to build those core capabilities. Since those two can be in limited supply, things can get tricky in practice. Also, since you actually need to build technology solutions to help implement the business strategy, things can get a little muddy as well. It's too easy for everyone to get distracted and "sucked in" to the current program. However, if the core issue is that technology people are trying to align with the wrong thing (strategy instead of vision), then it really doesn't matter what the technology is. It'll have the same characteristics over time as the business execution strategies are guided by, rather than aligned with, the core business vision. Maybe that's the real mistake. ast On Fri, 2008-05-16 at 18:30 +0000, Rob Eamon wrote: > So here is some food for thought. > > I was thinking about the success rate Gartner (or others) have in > predicting adoption/penetration rates for given approaches and > technologies. Since much of what is discussed in SOA has roots in the > EAI/B2B realm, I thought I'd try to see what the predictions were > back in the "heyday" of EAI. > > I was hoping to find a 1999/2000-ish article like "by 2005, 80% of > companies will have adopted an EAI strategy and toolset." Then, > ideally, find a 2005 article that indicated what the usage really is. > Alas, it seems noone cares about EAI anymore. It's a Bad Thing. We > only talk about the next thing now and that's SOA. That's what sells > newsletters, research papers and consulting. > > In my searches, I came across this Wiki page. > > http://en.wikipedia.org/wiki/Enterprise_application_integration > > I didn't find anything about adoption guesstimates but the "EAI > implementation pitfalls" and "Advantages and Disadvantages" sections > caught my eye. Could be "SOA" be substituted where "EAI" appears and > the sections still be accurate? > > One phrase that really caught my eye was "Most of these failures are > not due to the software itself or technical difficulties, but due to > management issues." Clearly all the chat about "SOA governance" > and "SOA lifecycle management" is attempt to address earlier stumbles. > > Have we really learned the lessons of the past? Or are we repeating > the same ol' mistakes? > > -Rob > > > ------------------------------------ > > Yahoo! Groups Links > > > -- Andrew S. Townley <[EMAIL PROTECTED]> http://atownley.org
