Service Oriented Architecture for Association for Systems Management (ASMTulsa.org)
I just delivered my first non-developer based presentation to the group Association for Systems Management.
It was very well received with a lot of great questions. One question that was brought up was, Why would any vendor want to implement an open standards interface like a Service? I believe the proof is in the pudding, just looking at the number of large software companies that have already built tools for SOA, etc. I can totally understand a big monopolistic software company trying to stay that way, but it's only a matter of time before customers demand the open interfaces or because of competition it becomes required. If they don't do it, then we as architects should build contracts around these closed systems, so that we can take integrate them in our SOA enterprise using a more autonomous methodology. The benefits of this architecture are tremendous. You can actually combine multiple back end applications into one contract if necessary. The example I used was a Financial institution and how they could have multiple back end systems for customer information. One main contract for each customer could easily work with multiple back end sub-systems.
I also believe a very good example of this form of adopotion across software vendors is XML. A few years ago very few software applications actually used XML. That very same debate was going on then. Look at how far XML has been integrated into everything.
It took some convincing, but I think I finally explained the benefits of this architecture.
One other interesting point that was brought up was "Why would developers want to hunt through a object catalog to find some previously built object versus just rebuilding their own? Most developers think it's just faster to write it again themselves."
Is this something that separates architects from developers? I personally think a developers main focus should be to provide an application that solves a need and by taking advantage of reusable code they can do so even faster. At least that's my opinion.
Comment Notification
If you would like to receive an email when updates are made to this post, please register here
Subscribe to this post's comments using
Comments
Leave a Comment
About dwalker
David Walker has over 15 years experience in application development with over 50% of that employed as a consultant with companies such as: Texaco, Bank of Oklahoma, Winner Communications (ESPN.com) and IBM Global Services. At the age of 14, he began his application development ambitions with a Commodore 64, BASIC, and a 300 baud modem. Even at that early age, he primarily focused on two specific application types: multi-user communities and database applications.
His hunger to learn as much as possible about development lead him through courses such as DBase III, DBase IV, Pascal, C, C++, Java, and several in UNIX. He started his development career first doing heavy processing with Access and VBA, then moved on to VB 3, Oracle, and Delphi. Visual Basic was one environment that remained constant for many years, including his very first .NET projects performed in Visual Basic.NET.
After working several years on very high end internal Corporate applications, the consultant company he was working for, sought out his ideas for actual software products that could be packaged and sold. He had already developed several prototypes of a dynamic portal application, before portals even became popular, so this became the logic decision and he became the Director of Product Development. Under his direction, a team of developers and graphic artists, took a skinning approach before that become popular, and completed the core portal application, and continued on to developer 15+ add-on modules, including things such as: Help Desk Ticket Systems, Change Control, Records Management, Human Resources, and many more applications. Eventually, it spun off into it's own separate company as KnowledgeGEAR, a complete intranet in the box solution.
Having worked as a consultant, he has had a experience with a very wide range of applications and architectures, at one time, even converting several Fox Pro and GW-Basic applications to VB 6 and ASP. His early training of Unix and the C language and years of experience with JavaScript, lead him very quickly to C#, where he has remained focused ever since.
He is the current President of the
Tulsa Developers .NET user group.. He has been an MCP since 2003 and MCAD and MCSD since 2005. He is currently pursuing his MCDBA and then on to MCSE.