Designing a framework seems like such an easy task, doesn’t it? Well, it is if you're the only one writing it. But how do you know you're writing a well-structured model for a framework? It seems that my version of a framework is not the version that someone else envisions. Besides, there are so many out there why would you even bother to spend the time to write one?
I can tell you why: When Microsoft wrote its framework they wrote in a manner that would be generic enough for anyone to implement.
Take for example ADO.NET. The wizards that are built into Visual Studio.NET allow you drag and drop connection objects, data adapter object and create strongly typed datasets to interact with your forms. All that works great…. If you’re working is Microsoft Databases.
What happens to those developers that are working in an environment where the business depends on the current database they have – even if the database is 20 years old?
Some companies believe in the saying, “If it isn’t broke, don’t fix it.” Except for when they begin working with technology that goes far beyond the scope of their technology.
Well, getting back to frameworks. Most framework manufacturers only create frameworks that are “cookie-cut” based on what Microsoft specified. They exploit the foundation by making use of methods which expose information about the underlying data structure. Those frameworks become useless when the underlying database doesn’t support those functions. The 3rd party frameworks fall apart!
So, we get stuck writing our own. One of the biggest dilemma’s were facing is architecture. Without good technical guidance you end up with a lot of disagreeing and discontent people. Ego’s begin to expose their bad side, people want their ideas to be implemented and management just watches as nothing gets done.
I hope this is not the case for the shop I’m working for – In the past two days of preparing for a framework we’ve had a lot of dialog going back and forth which is good. We’ve also had some well winded discussions that were needed in order to clear the muddy waters. I have to admit that I don’t feel we’re moving at the pace I’d like to see the framework move but we’re not dead stop in the mud, either.
