During free-test testing training, which I spend from time to time, each time there is a discussion about whether test descriptions need to be made and, if necessary, how detailed. I usually give quite radical advice - if you can, do not write at all, or if you really need to, write, but as little as possible. Write down not scenarios, but ideas, the result is a complete savings - first, less time is spent on writing, and then less time is spent on rewriting.
But saving time in writing is not all. Almost more important factor is that in the verbose descriptions the meaning is lost, which the test designer tried to lay there. Therefore, it is easier for an experienced tester to work with short descriptions than with detailed long scenarios. And today I want to bring to your attention the translation of a small note by Rob Lambert, in which he describes an experiment explaining this phenomenon.Less Is More, or Less Words - More Meaning.')
As an aspiring writer and a good communications specialist, I spend a lot of time learning how to improve the effectiveness of communications. And one of the best techniques is getting rid of those elements that are ignored by readers or listeners.
In writing, this means deleting words, sentences, or even whole pieces of text that you think, or you know, will be ignored. Of course, there is a lot of subjectivity. I also do not know for sure what can be thrown away and what should be left, but nevertheless, I subject my texts to hard self-editing.
If you use Twitter, you can imagine how this happens. Sometimes you have to work hard to be able to express your thought using only 140 characters, but the result is impressive. This is a very useful practice, because brevity, as you know, is the sister of talent.
Most of my texts become significantly shorter after going through the stage of “pressing”, although, if you try, you can still find a lot of cotton wool there.
However, now I want to talk about how this idea can be used to improve the quality of tests.
Let's take a look at the description of some test. Typical such a description, with a dozen or two steps and a bunch of small details. Yes, yes, these are common, many testers describe tests in great detail. I will even dare to declare that this is considered the norm.
There are several categories of people who will then have to read these descriptions. If you are targeting such testers who will blindly follow the instructions and put ticks and crosses - a detailed description is probably the right choice. The fact that this approach in itself is not very successful, we will not discuss here, if you choose it - this is your right.
Let's better see what happens if such detailed instructions fall into the hands of a
real tester . Even better, if you additionally ask him to give an assessment of how effective this test is. Here is what you see (if your idea of a
real tester is similar to mine):
- he will not follow the instructions step by step
- he will read the entire description to the very end in order to understand its essence and realize what this test is for
- he can highlight some areas to focus on
- he can read the test several times to double-check his conjecture as to what this test is intended for and make sure that he did not miss anything significant
- at the same time he will rely on all his experience and knowledge, and not only on the instruction itself
- he will perform the test, but not necessarily following the instructions step by step instead of following the steps of the instructions, he will first perform the entire test to the end, and then once again walk his eyes through all the steps and put ticks, controlling himself, did he forget to do something
- after that, he will probably try to perform several more similar tests, varying individual elements or parameters, exploring the surroundings, but not deviating too much from the purpose of the test (as he understood it)
- he will add to the test description his corrections, comments, suggestions, ideas
- he is likely to offer some new ideas for additional tests.
- and with all this, we emphasize once again - he will not follow the instructions step by step
Of course, these are just my own observations, but you can easily repeat them on yourself or on your colleagues. I'm pretty sure you'll see exactly the same thing.
Which of these can be concluded? Get rid of descriptions of tests from redundant text, irrelevant details, obvious things, verbal debris and other ignored elements. If testers ignore any suggestions, or ideas, or steps, or whole tests, just throw them away. After cleansing all the "cotton", you get what I call the "guide tests", that is, something like a checklist that
directs the actions of the tester, and not the governing document that
controls his actions.
This will force testers to include the brain in the testing process. And the brain is the most powerful and effective testing tool.