5 test myths

Recently I started to teach courses on the basics of testing, and it so happened that the group gathered from some programmers. And what was my surprise when people involved in software development, do not know anything about testing. Smart guys, advanced programmers, well versed in their subject area, do not know how to test the software they write.

Despite the fact that testing has existed for a long time, it is still a young and developing area, about which they know very little, outside the testing department. And then I wondered why people go to work as testers, or vice versa. Do any prejudices and myths move them? Somehow I came across an article by Mike Brown (Mike Brown) about 5 myths in testing , the translation of which I want to share with you :



Being engaged in the processing of test data, I am constantly amazed at how often I have to deal with cases of misunderstanding. There are quite a few misconceptions, but I would like to tell you about the five most common testing myths. The first three, it seems to me, prevail in news articles, while the other two are prevalent within the professional community.



Read and tell me whether you agree with me or not.

Myth one: testing is boring.

Often you hear that testing is like sex: if it is not fun, then you do it wrong. The myth of testing as a boring and monotonous activity is disseminated in the media, in which testers are viewed as working along the conveyor belt of the software industry. In fact, testing has to solve many new and interesting tasks daily. Here is a quote from Michael Bolton , summing up a peculiar summary of the above:

')

“In testing, we are driven by the desire to search for new information. Testing is a process of research, discovery, learning new and learning. When we set up, run and study a software product, in order to give it a rating or to discover some unexpected problem, we test it. We are testing, trying to delineate the boundaries and capabilities of the product and design. We are testing when we are driven by the desire to answer questions that are not answered or those questions that have not even been asked. "

Myth two: testing is easy.

It is also often said that software testing cannot be so difficult, since ordinary users constantly find errors in programs. In fact, testing is a very difficult skill, which the average person cannot master. Here is what Patrick Copeland from Google says about the required qualities of a tester:



“This is a special mindset and a special passion. On the 100 interviews I conducted for testers, I primarily paid attention to: 1) the presence of a special ability to search for problems and 2) the passion for testing that is present in a person along with this ability. In other words, testers love their work and do it well. They note that during testing one has to solve problems that are equal, and sometimes even surpass in complexity, those that have to be solved during programming . Anyone who is a tester from God and treats work properly will never be left without work. Such people are worth their weight in gold. „

The third myth: testers are just looking for mistakes.

Testers are really looking for errors, but this is not the only goal of their activities. The author speaks very well about this myth, having published an article under the name Ankur on the website freesoftwaretesting.info:



This is too limited a look at the work of a tester, which depreciates it in the eyes of users. Testers are experts of the system, application or product they are testing. Unlike developers who are responsible only for a particular function or component, testers are responsible for the operation of the system as a whole. Testers understand the importance of the product, the influence of the environment on its effectiveness. They know how to use the product with maximum efficiency.

Myth number four: cars will replace testers, and they will become unnecessary.

With the development of computer technology, it is becoming more and more common to say that testing programs by people will soon disappear altogether. But since the end users of computer programs are not robots or machines, but real people, the importance of testing performed by people will not diminish. Here is what James Whittaker, the author of books on testing, writes about the importance of manual testing:



With the help of test automation trying to solve too many problems. They are trying to charge too much on automation - and therefore it is so unreliable and vulnerable. Something can be done well with an automated system, but there is also something that people can do better. Therefore, a mixed approach seems to me more justified. Due to automation, I first of all want to ease my human work. Automation is needed to analyze data and identify models. But with the help of automated tools can not give estimates and make judgments. Fortunately, people do it better.

Myth Five: Testers do not get along with the developers.

It is understandable why this myth is so common. This was once written by testing guru James Bach:



Those who submit their work to the public must be prepared to discuss it. This is a very unpleasant feeling. And testers sometimes add fuel to the fire with hasty statements about the presence of certain defects in the program, posing as a quality defect that they just don’t personally like.



Not everyone knows that many testers are former developers (and vice versa), so they are quite able to understand each other and evaluate the challenges facing each other. Of course, this is not the case in all companies, but based on our experience, we can say that the rumors of misunderstanding between programmers and testers are greatly exaggerated.



I want to add to the list of myths: I heard such a myth from programmers that testers search for bugs only by pressing all the buttons in a row, without even thinking about the input and output parameters. There are also a number of myths about software automation, for example:

• automation reduces costs;
• automation solves the problem of lack of resources;
• the more autotests, the better;

Or the most common myth that testing improves the quality of a software product. To improve quality, you need to create it, and testing only checks the quality. Yes, of course, testers checked the quality of the software, found errors, showed a test report, programmers corrected, the quality improved. But non-testers improve software quality.



As for me, I went to testing, driven by some kind of premonition that there is room for development in testing, you will never stand still in testing. This development attracted me and attracts to this day. Although, of course, I thought testing was easy.



What testing myths do you know?





UPD. Thank you very much for your interest in my article. I will summarize a couple of myths / anti-myths / facts from Habr's readers:

Posted by: evans2094

Myth:

1. Developers and publishers listen to testers.

2. Product testing begins at the design stage and continues throughout its entire life cycle.

What is the development and testing reminds elephant wash. It can not be washed all cleanly and completely, while one ear is washed - the second one has already become smudged :) Of course, you can flood it with storm testing, but then the meticulous client will find some rubbish all over the elephant's surface ... and not completely soak in acid - after all :)

Posted by: netmike

Fact:

Automation software requires additional investment, although it can improve the quality of the product. In fact, automation begins to benefit only on products whose development and maintenance is quite long.

Adding from NatalyaRukol :

Automation is justified only in cases where the main costs are accounted for by clicks (stable, working functionality, which is already clear how to test and which does not contain a huge number of errors)

Posted by: graninas

Myth:

I write on haskell and I don’t have to test my code, everything works.

petropavel noted that these myths are universal and can be applied to any profession:

Myth one: cleaning the premises is boring.

The myth of cleaning the premises as a boring and monotonous activity is disseminated in the media, in which cleaners are seen as laborers of office activities. In fact, the cleaning has to solve many new and interesting tasks every day. [approx. trans. for example, chewing gum on a chair]



Myth two: cleaning is easy.

It is also often said that cleaning cannot be so difficult, since ordinary

people constantly clean their apartments ...



The third myth: cleaners just pick up trash.

Cleaners do pick up the garbage, but this is not the only purpose of their activities ...



The fourth myth: cars will replace cleaners, and they will become unnecessary.

With the development of computer technology, it is becoming more and more common to think that cleaning people will disappear altogether soon ...



Myth Five: cleaners do not get along with office workers.

It is understandable why this myth is so common ... Not everyone knows that many cleaners are former office workers ...

wicked_sten refuted one of my myths that testers do not improve software quality by writing:

The number of my innovations is slightly less than requests from customers. Well, somehow it happened. The simplest are done right away, complex ones - after a couple more requests for the same.

Ps. Thanks a lot for the translation to my colleague AndreiYemelianov