Unit Tests are a good thing.
I concur with that statement! I must admit, I have been half hearted with my unit test attempts in the past. Mainly because I haven't put the time into arranging the pieces to make it happen. But with my last project, one that involves multi-threading, I went in with both feet. Do you know what I found? I found I was able to move along much faster. I felt confident with my code. The code was better organized because when I wrote my tests the organization of code was centered around "units". I used NUnit 2.0 and setup my VS.NET project with two targets one that ran the unit test and the other that ran my code in user mode. I'm on my way to finishing this project but I can say that I will be using unit tests a lot more in the future.
[StronglyTyped - Richard Caetano's weblog on software development]
The problem has always been convincing people that unit tests do more than they sound like they do. It's hard enough convincing people to spend the time devoted to developing them in the first place since I think you need to have that first moment of relevation where you realise that your tests just saved you a great deal of time. This seems to be through either a refactoring or realising that the code you thought would work actually has had a great deal of things wrong with it that you wouldn't have realised until the bug reports started coming in. Not everybody here is convinced of their real value yet and it's very difficult to help them to realise what an advantage they are. You really have to reach that conclusion on your own.
What other blogs are saying about this post.
10:27:22 AM
|