Forrester Research: Dynamic Business Applications, SOA, Web 2.0
The Forrester IT Forum 2008 in Orlando has come and gone. It was a very informative event in difference to many others. I am not yet sure about the marketing value for a software vendor and what projects might come from it. Time will tell if the right people were there.
I went to listen to a number of sessions, spoke to a few participants and tried to connect with some more analysts. I could not get hold of Kyle McNabb, but after a phone tag game with Connie Moore she found me sitting there a bit forlorn. I had one key question for her: ‘How do you see the current rigid flowchart BPM model evolve, especially in the light of content and rules integration?”
Connie Moore’s answer suprised me: ‘I don’t believe in rigid process models.’ So we ended up discussing the Forrester concept of Dynamic Business Applications that focuses on ‘Design For People. Build For Change.’ I feel that DBA is very close to what we offer with the Papyrus Platform. She said that most BPM vendors are unhappy with Forrester’s segmentation of the BPM market into integration, document and human centric process management systems, but that at least gives businesses some way to understand what a system is good at. I can go along with that as much as I do not like the market segmentations myself. She said Papyrus could be rated as a document centric process system with our UTA process training and the Web 2.0 capabilities as clearly distinguishing factors to the others. Makes sense too. She said that we are not a typcial BPM system if we do not focus on flowcharting. I totally agree there. We once again created our own marketplace that we can not be compared in. That actually was a relief to me. On the other hand Papyrus is much better suited to adress process issues than your typical BPM product, mostly because of the powerful content, GUI and integration features. But how will we make that understood?
The other two very prominent subjects in Orlando were SOA and Web 2.0. IBM’s Sandy Carter offered her book ‘The New Language of Business – SOA and Web 2.0’ published by IBM Press(!). I did get a signed copy and read it on the flight back from Orlando to Frankfurt. Well, she has some good points in there but I think the title is misleading. There is no ‘language’ aspect to neither SOA nor Web 2.0 and it certainly is not a language that anyone involved in business will understand. That is the reason why Sandy clearly recommends that you need a strong partner (IBM obviously) to make it work. Sandy is clever enough not to use ‘agility’ but creates a new word ‘flex-pon-sive’ as a mix of flexible and responsive. In the end she offers nothing new or a true insight as to how to combine SOA and Web 2.0.
So we are back to the problem of a huge gap between what the business needs and how they can describe what it is. SOA and Web 2.0 in their current forms ARE NOT solutions that address that. They are technical concepts that may play a part in actually creating a total solution but require a huge manpower investment. One thing shocked me though. IBM had on their booth a big sign, claiming that they have 700 (yes, SEVEN HUNDRED) patents related to SOA. Which means that if YOU are involved in SOA today you are already infringing and liable to lawsuits. I am sure that Microsoft and Oracle have as least as many patents. Another reason why I think that SOA as a technology is DOA despite its good concepts. If I were you, I would stay away from SOA and do something for my users!
A related subject I had discussed with Forrester’s Mike Gilpin, namely ‘federated SOA’. As far as I understand it, it relates to the problem that businesses are unable to create a single SOA infrastructure and thus need to federate SOA interfaces. No surprise there. I had a similar discussion with Dr. Dorn of the Technical University Vienna who was the mentor for a study project into our User Trained Agent. He said that the main purpose of SOA was linking backend services together and not a single homogenous infrastructure.
Right! If SOA is not only about WebServices and not about a single homogenuous infrastructure (that should handle GUI and process) then SOA ends up being a very broad term for loosely coupled service interfaces. What a waste of time and money all the advertising and buzz is.
Dear business user: Do you know who is paying for all that? YOU ARE! Nearly 50% of most IT companies revenue go into ‘cost of sales’. You get nothing for that at all!