• Visitors can check out the Forum FAQ by clicking this link. You have to register before you can post: click the REGISTER link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. View our Forum Privacy Policy.
  • Want to receive the latest contracting news and advice straight to your inbox? Sign up to the ContractorUK newsletter here. Every sign up will also be entered into a draw to WIN £100 Amazon vouchers!

Service Oriented Architecture

Collapse
X
  •  
  • Filter
  • Time
  • Show
Clear All
new posts

    Service Oriented Architecture

    Anyone know a url with a clear, succinct definition of this, i.e. one that's easy to regurgitate in an interview? I can honestly say that having worked with WCF I still can't explain what SOA has over DCOM or a bunch of windows / web services in a non SOA set up.

    #2
    If it had a definition there would be no fun (i.e. fat consultancy fees with no visible outcome). Have a look at this, and its follow-up, for some ideas and sympathy:

    http://martinfowler.com/bliki/Servic...Ambiguity.html

    Comment


      #3
      Interesting, just seems to confirm my cynical view that it's all about recycling all the buzzwords and soundbites that were associated with DCOM. Sure, web services are more loosely coupled than DCOM but from what I understand just putting your implementations behind a web service interface doesn't in itself make it SOA.

      Comment


        #4
        I think you need to abstract it from individual technologies. I'd say DCOM could be a valid way of implementing an SOA, though web services are more common. For me SOA is more about intention than implementation.

        Comment


          #5
          SOA is much more about the business than the technology backing it.

          You are right, loose coupling is a core concept, but the reasons why and the benefits to gain from doing so across the enterprise is where the value comes from.

          SOA isn't about webservices or technology platforms, unfortunately most people think it is and hence why people have started to go on about SOA being dead.

          My 2 pence worth
          TM

          Comment


            #6
            Originally posted by thunderlizard View Post
            For me SOA is more about intention than implementation.
            Good interview sound-bite
            Originally posted by MaryPoppins
            I'd still not breastfeed a nazi
            Originally posted by vetran
            Urine is quite nourishing

            Comment


              #7
              Reusable stateless units of functionality that can be accessed over http across corporate firewalls blah blah.

              Comment


                #8
                Originally posted by thunderlizard View Post
                I think you need to abstract it from individual technologies. I'd say DCOM could be a valid way of implementing an SOA, though web services are more common. For me SOA is more about intention than implementation.
                Originally posted by themistry View Post
                SOA is much more about the business than the technology backing it.

                You are right, loose coupling is a core concept, but the reasons why and the benefits to gain from doing so across the enterprise is where the value comes from.

                SOA isn't about webservices or technology platforms, unfortunately most people think it is and hence why people have started to go on about SOA being dead.

                My 2 pence worth
                TM
                Originally posted by Diestl View Post
                Reusable stateless units of functionality that can be accessed over http across corporate firewalls blah blah.
                You all sound exactly like the examples given in the article (and the followup linked therefrom) that thunderlizard linked to

                I think this proves that Fowler and Ing were correct - the term is basically meaningless

                Comment


                  #9
                  Originally posted by NickFitz View Post
                  I think this proves that Fowler and Ing were correct - the term is basically meaningless
                  Fowler is an arse on naming conventions "It's not 'Inversion of Control' it's 'Dependency Injection' because I said so"

                  Most of us just called it priming the object for years.

                  Software would be far better off without fowler's opinions.

                  Comment


                    #10
                    Just because a term has a number of related meanings, rather than one single unified meaning, that doesn't make it meaningless.

                    Comment

                    Working...
                    X