Eric Jan Malotaux presents a way of implementing SOA in small iterative steps, each step delivering value to stakeholders, higher than the costs involved, using feedback from previous steps to adjust the requirements and design accordingly.
Eric Jan Malotaux has been developing software for more than 25 years, using many different programming styles and programming languages. Currently he works for Ordina, a large ICT service provider in the Netherlands. He specializes in software architecture, development environments, model-driven development, and agile development processes like Evo, Scrum and Extreme Programming.
The International SOA Symposium is a yearly event that features the top SOA experts and authors from around the world, providing a series of keynotes, talks, demonstrations, panels, and SOA training and certification workshops - all with an emphasis on realizing SOA in the real world.
Re: Could not download the slides.
Thanks a lot for letting us know. The problem has been fixed.
Tor Arne Kvaløy
Re: Real understanding
I am not quite sure where you see any particular knowledge about SOA in this presentation. Somebody who does not understand (let alone speak about) the relationship between reuse and service versioning has no understanding whatsoever about SOA. Somebody who does not understand that one of the key principle of SOA is the decoupling of the interface from the implementation (something that REST 99% of the REST developer don't practice) and claim that an ESB has no value, has never found SOA. You don't have to spend millions to achieve this separation, but not understanding how core it is, is very sad.
The author may think that he speaks about SOA but he doesn't. He may have achieved building some "usable" software apply EVO principles, but by a large margin he has never build a "service". The author points rightfully that "reuse" couples service consumers, but he has no understanding on how to use SOA technologies to minimize this coupling and therefore rip the benefits behind SOA. This is why the author claims that "reuse" is "overrated".
For the record, "reuse" doesn't have to be designed. It can't be designed. In SOA, reuse happens the other way around, it is not new consumers who reuse existing services, it is the old consumers who reuse new service versions (without any change) as the service evolve to meet the needs of new consumers. Anybody who is trying to design a service that can be reused 3 years from today is wasting everybody's time and resources. Governance does not yield reuse.
It is sad that in 2011, people still propagate this kind of deeply to and totally flawed message about SOA.