Why Does Hazel Love Testing So Much?
*Updated with bio info 8/24/09
As part of an interview process, I sometimes get asked the question “So why test, what makes testing your primary choice in careers?” Why not be a software design engineer, what makes test more attractive.
The fact is I am pretty good as a developer, I also like leveraging what creativity I have, and I enjoy seeing the fruits of my labor. But I get that in test just as well as I would if I were developing a product, and the fact is, that is a core part of my test deliverables.
But what I really like, what really drives me is something that I think very few people would see in my job. To me, testing is absolutely the embodiment of the scientific method. It is as much tinkering and experimentation as it is software development. It is enumerating some assertions (test cases) and then applying your methodology (developing automation or just pushing the buttons) and seeing what happens.
I’ve been doing this for so long, that as soon as I go into a shop, game store, hobbyist shop, that I already have the top 3 or 4 things I am going to look for to assess whether I am getting a quality product. I can visually see if the product is of quality just by looking at the box. Does the grammar make sense, can I find all the product visual in one look, does the packaging reflect the product, are there flaws on the package from poor shipping or storage, and is the product warranty clear? Finally, when I purchase the item, I take it home and start pushing its buttons. If it doesn’t perform the function I expected, I return it.
That is easy for me and you would think that I am doing what any good shopper does. But then, isn’t that the point? Shouldn’t my inspection of software start with the wrapping and then work my way into the package? When I dig into software, I start at the functional specification, then the design, then the feature itself, and I ask the software equivalent of those same questions. I get answers in terms of a final specification, and then I build my list of test cases and test scenarios, automate what I can because I am lazy, and push the buttons on what I could not automate, and exercise the scenarios. If the feature/product does not work the way I expect since I was part of the specification process, I reject it without prejudice. In reality, this is a churning process whereby I am opening and someone is fixing bugs. The end result is usually something I am proud to have worked on and willingly allow to go to market.
So, why test? Because I like experimentation and I always have; assume nothing is perfect, and prove it through rigorous experimentation.
Technorati Tags: XPe,Standard 2009
Comments
- Anonymous
August 23, 2009
Test is very vital to confirm something does work. Thats what i learned at school and university ;)