8 Replies Latest reply on Oct 5, 2005 11:47 AM by mutpup

    New Application Dev Thoughts

    mutpup

      I find myself in a rather unique situation that allows me some different options that I can go with.

      I'm just starting on a new application that I wish to build. My previous experience is basically with writing JSP pages with backend classes what make connections and execute raw specific sql select and updates. So not much if any expereince with EJB's but I know the basic concepts.

      So the question becomes if you had to start learning again, and could choose your tools how would you develop a new app.

      This doesn't have to be production until Febuary so I can start with beta tools knowing they will become standards before I go live.

      Is Seam the way to go?

        • 1. Re: New Application Dev Thoughts
          rdewell

          IMHO, the following in your toolbelt will make life wonderful:

          Java 5
          EJB3 for persistence, etc.
          Facelets
          Seam

          Ryan

          • 2. Re: New Application Dev Thoughts
            rdewell

            You also might consider using a workflow solution from the beginning. I.E. something like jbpm, which Seam integrates with.

            Ryan

            • 3. Re: New Application Dev Thoughts
              gavin.king

              Absolutely Seam is the way to go!

              What would you expect me to say ;)

              • 4. Re: New Application Dev Thoughts
                lavoir

                Let me try and be a little more objective then all the Seam fans here ... :)

                Seam is the way to go :P

                Seam is only a part of the puzzle, JSF, Facelets, Hibernate, EJB3, etc .. are also part of the picture. Therefor, your not just learning Seam, you are learning all the new JEE standards wich is desirable regardless of wheter you end up using Seam or not.

                • 5. Re: New Application Dev Thoughts
                  mutpup

                  I understand the excitement level, but do you see seam as the way things will be going over the next 18 months? EJB3 is definitly on it's way, I'm wondering if you feel seam will start to gather that same momentum?

                  • 6. Re: New Application Dev Thoughts
                    lavoir

                     

                    "mutpup" wrote:
                    I understand the excitement level, but do you see seam as the way things will be going over the next 18 months? EJB3 is definitly on it's way, I'm wondering if you feel seam will start to gather that same momentum?


                    I personnaly think so.

                    But since your using java standards, nothing stops you from pulling out Seam from your app and writing your own glue code later with little change in your existing code if you ever decide Seam is not for you.








                    • 7. Re: New Application Dev Thoughts

                      IMHO, long term, the only thing you can depend on is your POJO objects you create. That's it. So selecting technologies like EJB 3, JSF, and Seam-- where they are built around working with POJOs, you really don't waste time writing integration code for frameworks that may change.

                      For example, looking back at a larger project of ours, it seemed like half our code was in Struts Forms and Actions, where the above technologies 'just work' with other half of your code that you actually care about.

                      -- Jacob

                      • 8. Re: New Application Dev Thoughts
                        mutpup

                         

                        "lavoir" wrote:


                        I personnaly think so.

                        But since your using java standards, nothing stops you from pulling out Seam from your app and writing your own glue code later with little change in your existing code if you ever decide Seam is not for you.



                        Right, that's how I'm trying to view it with the layered approach. Worst case I have to ditch a layer and rewrite a small portion of the code.