Skip to Content

last night I read an interesting blog about the present situation and missunderstandigs of SOA by Anne Thomas Manes, a well known Enterprise Architect and author of a Manager’s Guide to WebSerrvices http://www.bowlight.net/webservicesbook/index.html

 

She postulates “SOA is Dead – Long Live Services” and by the way created a SOAsaurus, highly influenced by the economy.

Her headline might be very confusing at the beginning but while you’re reading and understanding her argument and her view about what’s needed to do “SOA”, making “SOA” and want to establish “SOA” in an Enterprise Architecture she points out some provoking opinions but mentioned useful key points.

 

Her first painful statement postulates:

“It’s time to accept reality. SOA fatigue has turned into SOA disillusionment. Business people no longer believe that SOA will deliver spectacular benefits. “SOA” has become a bad word. It must be removed from our vocabulary […] But perhaps that’s the challenge: The acronym got in the way. People forgot what SOA stands for. They were too wrapped up in silly technology debates (e.g., “what’s the best ESB?” or “WS-* vs. REST”), and they missed the important stuff: architecture and services.”

This might be confusing but I made experiences with IT people either smiling when the hear about “SOA” strategies, be very reserved about the costs (especially mid market customers) or still doesn’t really understands what SOA really stands for? She points out that SOA DOES NOT stands for a technology which could be bought by a customer. Often it is still not accepted that SOA is a “mentality” which uses a “mindset” and a set of “patterns” to achieve an architectural redesign which ends up in “IT efficiency” and “Agility”. The way to achieve this is still hard work and painful but with the right strategy, including the right technologies it’s manageable.

 

“And that’s where we need to concentrate from this point forward: Services.”

Her last sentence is (in my opinion) the key points which claim what we really need to concentrate on most when we’re talking about SOA and Architectural Enterprise Design -> its Services !!

This thinking needs on the other hand a clear and definite specification what a “Service” in your specific organisation really is, stands for and what this service needs to fullfill. So her blog is actually another hymne to what we really need to concentrate on: Service Governance with its Service Design and its Lifecylemanagment.

 

Please be encoured to read Anne Thomas Manes blog if you are interested in EA and SOA topics

http://tinyurl.com/9gdyer

 

An interesting comment on her blog can be found here from Mike Kavis http://tinyurl.com/7nzn45

What Do you think about her provocing blog ?

 

——- Additional comment 2009/01/30 ———

Maybe the term “Agile Enterprise Architecture – Agile EA” has to be used instead of “SOA”

http://agileea.com/portal/index.php

SOA is definitely not dead !! It is more alive than ever as Agile EA !

——————————————

To report this post you need to login first.

7 Comments

You must be Logged on to comment or reply to a post.

  1. Bernhard Escherich
    Hello Dries,

    thanks for sharing these very interesting links. They are very thought-provoking.

    But I disagree on several points:
    1. SOA is not dead. Especially in my industries (public sector and healthcare) SOA is widespread and without a real alternative for the years to come. Just think about public tenders where SOA as a concept is very important.

    2. Even if you concentrate on services- what´s the point? Should these services be deployed without an architecture? I agree that the task is a huge one but to declare SOA dead is not very helpfull. Instead it seems to be a normal position now on the Gartner hype cyle (disillusionement).

    3. I missed in your post and the links completly the connection with business processes. That is the real challenge to understand them, redesign them and then work out the different layers of an architecture. Without these foundation it is just a technical architecture and too shorthanded.

    I would be very interested to get a feedback from you if your experiences as an enterprise architect support my last point or if you have made other experiences.

    Best regards,

    Bernhard

    (0) 
    1. Dries Guth Post author
      Hi Bernhard,

      many thanks for your additional comments because you mentioned an important issues this blog didnt’t emphasized: Business Processes! This might be my fault but wasn’t my focus.

      to 1.) I wanted to point into the direction some architects are thinking nowadays. One might think Anne Thomas Manes is denouncing SOA but she really wants to reach the opposite. By declaring the word “SOA” as a dead word the hype about SOA ends up and we can now concentate on service oriented architecturing and redesign landscape & businesses without any concern about this wonderfull strategy.

      to 2.) I am convinced Service Design is still the key to service orientation besides using the right technology/architecture (wich have to be able to support the service orientation) governed by Business Processes. I think we are still in a phase where we need to concentrate on this topic of Service Design to create Services which stands for its own, well designed, fine granual and support all features a service needs to support (theres a wonderfull book by Nicolai Josuttis out there http://www.amazon.de/SOA-amp-IT-Management/lm/R22EAKAV2J5ON7) and created business value. These services MUST be supported, or better used or mirror the right business functionalities and have to be implemented in the right manner. So I still be sure that the key might be the Service which will be (from TOP) be designed/governed by the business (Process) and (from BOTTOM) be support by technology. All the needed redesign of architecture needs to be considered too by the architect.

      to 3.) For myself I have still some concern and made these experiences: “SOA” is still not really understood by a couple of people, sometimes missunderstood, sometimes missused or sometimes even used to hype their own topics. Sometimes from a too technical perspective, sometimes from a too BP modelled level, and sometimes only from a Business Perspective (even this is in my optinion only the one side of the coin). Especially in the mid market sector I am mainly focused on “SOA” is sometimes still a disillusionement but FOR SURE the strategy mostly everybody wants to adapt to. I am still thinking for myself coming from a too technical perspective and I often catch myself thinking in a too technical dimension. In one of my current projects we’re started with a full business process modelling phase before we started to design the architecture and the “real” processes. The term “redesign” or even “SOA” was never used (by the customer) but the customer is willing to go a new architectural way governed by it’s new and common business processes we’ve modelled. So we used “approaches” from a SOA strategy but I would not call it a Service oriented Architecture because we DO NOT design well defined granular services. The Business Case (without focusing on Technical Services) is the goal and therefor I really agree in your last point.

      (0) 
    2. Thomas Obermeyer
      Thanks Bernhard for pointing this out.

      Also in the banking industry we are experiencing a severe need for SOA based design principles and solution approaches. Only when starting from a business process decomposition and a proper service modeling methodology SOA makes any sense. That’s also what I have mentioned in my blogs on SDN (Introduction to Service Oriented Architecture in Banking).
      Unfortunately, SOA has often been driven by the IT departments and had a strong focus in the technology aspect. I believe, this is the reason why some analysts now start to declare SOA dead.
       
      Best regards,
      Thomas

      (0) 
  2. Julian Reschke
    “SOA” already was a silly term in the first place.

    Replacing “Service Oriented Architecture” by “Services” won’t help either.

    Just my two cents,

    Julian

    (0) 
  3. Paolo Romano
    …that SOA is not a ‘must’, but an opportunity.
    I use it in everyday’s working task, when designing and developing new interfaces as well as when I think new applications

    Think soa will grow in silence, without be noticed because the change in mentality of new architects, developers and all the ‘workers’ that have the vision of the advantages. SOA is not a ‘big bang’, but a slow process that will take from dino’s to future.

    (0) 
  4. Anonymous
    Hello Dries,

    I cannot agree more with you regarding the approach taken to represent SOA. From the days of its inception the concept of SOA was mistaken to technologies and often the vendors supporting SOA methodology put forward a bunch of tools and toys instead of really aligning with the business needs and business jargon.

    The concept of SOA has been in use in principle for thousands of years evolving with the evolution of man on the face of the earth. We as individuals use a lot of services like postal services, pay services or tool services, where we don’t want to be bothered with the working details of these entities but would just like to use the services to accomplish the needed activity. Is this not activities based on services? In fact this is the under penning of the SOA methodology, where the user should be only concerned with the usage of a service but not with the actual workings of the service implementation.

    The concept of SOA is engrained into every part of our day to day life, but the representation of SOA in the IS space is really having identity crisis. This is due to the fact that the vendors choose to milk the SOA cow with a lot of tools at the expense of both the customers and the methodology.

    Do we need SOA? the answer is YES, but like you mentioned may be its time the word is rebranded as there is a lot of baggage now hanging around.

    Best Regards,
    Naveen

    (0) 
  5. Faycal CHRAIBI
    Hello Dries,

    I have published recently a blog post on the same theme, reacting to Anne’s original post.

    You may read it here: The specified item was not found.

    In the whole, I do agree with you. But again, you need to have a correct definition of what a service is. This might as well be confusing as the rest of the SOA concept. The Service definition has always been one of the main SOA challenges. It is very dependant of the project scope, the industry, its practices/processes and business objects (aka manipulated data entities).

    (0) 

Leave a Reply