Geeks With Blogs

The Lanham Factor The (ir)rational thoughts of a (not-so)mad man

My good friend Brian said something very interesting to me last week that has stuck with me.  We were talking about using the Microsoft Enterprise Library for data access.  His comment (and I'm paraphrasing) went something like... "Are you sure you want to use that piece of software wherein who knows if connections have been closed, objects have been disposed, etc. etc.?" 

I'm one for buying/downloading if appropriate as opposed to building.  However, I totally respect Brian and his perspectives.  As a result, this one intrigued me.  I spent a lot of time thinking on it and came up with what I think is a good reason to use an open source product, such as the Microsoft Enterprise Library as opposed to building your own data access components.  It so happens that my reason also describes a flaw in his logic.

His logic is flawed in the assumption that custom code will close all connections, and dispose of all objects.  Or that if that doesn't happen, a code review will catch those situations. are truly one of the GREAT developers.  You are conscientious, professional, and thoughtful in your code.  Therefore, I have no doubt that you could write an equivalent Enterprise Library (and, if I recall correctly, have done so in some fashion). 

Consider most coding efforts and the level of attention given to quality control, code cleanup, etc. etc. (as opposed to the "jist git r dun" mentality).  You may quickly realize that having a community reviewing the code instead of a one or two developers is actually better.  As a case in point, consider an article just released on testing the Microsoft Enterprise Library (DDJ article).  This article shows what an open source community can do to help. 

Sure it's not perfect.  Sure great developers like my friend Brian could do (arguably) just as good a job if not better.  But when you consider the time to market pressures associated with ubiquitous corner-cutting in QA, I think that using these types of components should be considered heavily.  At least consider them...

Posted on Tuesday, February 20, 2007 3:00 PM Cutting Code | Back to top

Comments on this post: Open Source COTS Components vs. Custom Compnents

No comments posted yet.
Your comment:
 (will show your gravatar)

Copyright © Brian Lanham | Powered by: