My blog this week is based upon this article in which the author explores “shift left testing” which is simply a fancy way of testing earliest in the production of a program as possible. In this article the author goes into detail about several different forms of shift left testing, all of which try to push the testing earlier into the development of a program. The author also reviews why the common practice of testing late in production phase has been a major harm to system and software testing.
I found it interesting that the testing phase is often left so late in the development, with some major consequences pointed out by the author. One of these consequences the author highlights about this method is that it often results in “wasted time”. This can happen because “Many requirements, architecture, and design defects are not uncovered and fixed until after significant effort has been wasted on their implementation.” This article also focuses on what I feel is a much better approach to the development in which each part of the original “V-model” of production is broken into separate “V-models”. This means that for every part of the production phase, whether it be design, coding, fabrication, it is accompanied by a testing phase. This does require more time in the production but will ultimately produce a better program with fewer to no faults. I believe employing this strategy in one’s own personal work would be a very smart idea. This can be done for example while creating a program yourself, with every part of the program you write, you test it to make sure it is correct. This will help by the end of the program when you could generally have several mistakes causing errors throughout the program. One final thing the author points out which I believe could be relevant to those not employing this method is that late testing can lead to missed deadlines, whether it be for a company or for a student working on a project.