I caught Beth Gold-Bernstein's wrap-up of Software AG's Innovation 2008. I always enjoy speaking with Beth, but this portion of the post needed to be addressed here:
"Dave Linthicum, always the provocateur, put forth the vision that angry masses were waiting at the gates with to torches, because it's too difficult to implement."
That's actually not the point I was attempting to make. My core assertion was that the technology is not doing a good job in matching up with the requirements of the typical SOA practitioner, generally speaking. In essence, what I'm hearing, is that the enterprise SOA requirements are much more rudimentary, and the existing SOA technology -- while innovative in some instances -- just does not match up to the needs.
The fact of the matter is that when you sell technology you're looking to be as innovative as possible, and thus love to drive ahead of the market. I've done that many times in my career. While innovation is indeed a desirable characteristics of technology, there are many areas where the SOA technology has a tendency to drive to innovation, while at the same time missing the needs of the rank-and-file SOA practitioners.
While the SOA vendors are quick to point out their success stories through customer case studies, that does not mean I'm wrong here -- only that somebody found the technology of value. Again, I'm talking about the larger general marketplace and the core requirement of the typical enterprise attempting to make SOA work for the first time. The technology is just not lining up well with the requirements, according to the SOA architects I'm speaking with.
The best example that I can find of this same mistake being made is the release of Windows Vista, as related to the forthcoming release of Windows 7. Windows Vista just had too much going on, and many of the features and functions just were not needed by the typical business and home user. Thus, they remained with Windows XP, and the pushback on Vista is now legendary.
Windows 7, if I understand what came out of the recent Microsoft PDC, is going to be much more componentized and "lighter weight," even able to run well on an underpowered $500 notebook with 1GB of memory. Also, they will be removing some of the features that added to the overhead of Vista and that users found bothersome, me included.
Indeed, Apple is moving in similar directions with the release of its next operating system code named Snow Leopard. In essence they are adding no features, just making the operating system faster and more efficient. Apple is listening to their customers, and now Microsoft is as well.
Before I get a bunch of comments here about operating systems being very different than SOA technology, I clearly understand that. My point is that sometimes you can put too much into technology, attempting to differentiate and be innovative, but not meet the needs of the intended users. I believe that's the case with some of the SOA technology being sold out there, and it's time to learn to align to needs. You'll sell more, believe me.