tag:blogger.com,1999:blog-13831777.post-62059063457676192102007-12-21T23:35:00.000-05:002007-12-21T23:35:00.000-05:00I agree with you in some of the points you expose, but I disagree in the overall idea. You make very good points when you say, <I>"Testing itself does not produce quality. There are only two things that produce quality: the process that you use to create the product itself (not including testing), and the responses taken based on the knowledge gained from testing."</I> Therefore, you want to gather information from testing as soon as possible, and the quickest way is to test.<BR/><BR/>Of course, you cannot possibly think of every scenario your customers are going to run into (or I should say, it is not likely you will be able to define every possible scenario). Under this assumption, releasing often will help you to obtain feedback from your actual customers very soon, which will allow you to deliver fixes to them very quickly. This will increase your software quality, but not from the fact that you release often, but the fact that your customers are testing the scenarios you have not thought about.<BR/><BR/>You said, <I>"Let’s say you produce a product with absolutely no test plan or you have a great test plan but you don't execute it. There is a chance that the quality that the customer experiences will be exactly the same as if you had a test plan and executed it."</I> Although this is statistically possible, you will have to agree that the actual chances of something like that happening are very slim. If you don't have a test plan, or you have one and don't execute it, it is more likely your final product release will have lower quality than if you executed the plan (no matter how bad the plan was) because you won't have any feedback until your customers actually uses the product. I am sure with your experience you are not trying to convince us that we should release our products without any actual testing. You probably know as well as I do, that it is not a good idea to release untested software, especially if you want to keep your customers around.<BR/><BR/>The moral of this story and where I disagree with you is that releasing often does not increase quality. Good processes and testing often increase quality. Good processes will allow you to make less mistakes. Testing often will help you to find the little mistakes you make, because we all make mistakes, quicker. With that information, you can act and fix the issues found. Releasing often will help you to gain feedback from your cutomers and act upon what they report, but releasing often will not help you if your often released software is not used; therefore, releasing often does not improve quality; testing often does.Isaac Rodriguezhttp://www.blogger.com/profile/04341790584083395302noreply@blogger.com