Tuesday, 14 December 2010

Sour Grapes

People who follow my blog may already know that I have an interest in psychology and how it can impact our thinking in relation to testing. Previously I have written articles on emotions , feelings and confirmation bias. I intend within this article to look at Cognitive Dissonance and how much of a help or a hindrance it can be for a tester.

There have been many studies on Cognitive Dissonance and if it really does exist however it would be best to describe what it is first. Wikipedia describes in it very simple terms:

Cognitive dissonance is an uncomfortable feeling caused by holding conflicting ideas simultaneously.


The history of Cognitive Dissonance is accredited to Aesop’s tale of the Fox and the Grapes where the fox who wants to eat the grapes cannot reach them and as such to remedy the conflict that he really does want the grapes decides that the grapes are either not ripe or too sour, hence the title of this blog.

My thoughts on this conflict of beliefs or opinions and how we adjust to resolve the conflict is something that I feel could benefit testers especially when carrying out exploratory testing.

As testers I think it is important we recognise when we are experiencing Cognitive Dissonance and learn to ask questions before we make a decision to resolve the conflict. The problem with experiencing cognitive dissonance is that it is very easy to change our opinion and belief and suddenly we are drawn into the trap of confirmation bias.

As humans we do not like the feeling of conflict within our minds and we try to make a decision to resolve this conflict. Once we have made that decision we will try to justify the reasoning for making the choice we did.

For example if you had two areas to test and each area had the same level of important and you choose area A instead of B. You will now subconsciously favour choice A more than choice B. Now if someone comes up to you and says Area B is more important you have a conflict and your mind needs to give reasons as to why you choose A. You may make statements such as well at least Area A has been tested, it was important for me to test it. You will try to justify the reasoning for making what appears in your mind is the wrong choice.

When testing if we get two conflicting oracles and you need to make a decision on what is correct or the product owner needs to make a decision, cognitive dissonance could come into play. A decision could be made which is wrong. Only the problem now comes at a later date when you need to justify why you made that decision you will change your opinion or belief to make sure that the decision you made was the correct one.

It can become worse if you use a rating system to rate identical items. Imagine you are in a team and the team is given the task to rate features to be tested or developed. These ratings are then used to determine the order in which features for the product are developed. You then make a choice to test/develop your most highly rated feature, at a later date you get to rate the list again the item that you rated as next best has suddenly become of low importance even though it has the same value as the previous highly rated item. Making a choice after rating affects your value/belief/opinion of the item of same identical value and you start to score it lower to stop your uneasy feeling or cognitive dissonance. Project manager and team leads need to be aware of this since something of high importance could be downgraded by a team because of the issues of cognitive dissonance.

So how else can this affect testing?

I believe it can actually harm what we are testing. If the belief we have for the software we are testing in what it should be doing contradicts what it is doing and we adjust the conflict to justify what the software is doing there is a danger we could miss an important bug.

Further reading:


This is going to be my last post before the holiday season so I would like to say to all readers old and new have a happy holiday and look forward to writing again in the New Year.

Thursday, 9 December 2010

Sorting the chaff from the wheat

One of the questions posed at Eurostar 2010 hot topics panel session (link) was:

What’s the most important skill for a tester?

Michael Bolton (Link) gave a witty reply of

Recognizing there is no most important skill.

Whilst I agree with Michael that testers need a wide variety of skills it got me thinking about skills that all testers need or should have. One of these is being able to deal with the vast amount of data that everyone has to deal with. How do we deal with?

I have recently been re-reading the classic HG Wells story “The Time Machine (Link) and started to think about the two communities with the story, the Eloi and the Morlocks and how society in general is getting so much information that it is starting to make us dumb, this IMO is a dangerous thing for testers.

Are we as a society becoming like the Eloi?

They have access to amazing technology that helps with all their needs however they come across in the book as dumb and lack curiosity. They see no need to be thinkers or philosophers. It appears technology is making things easier and easier for us.

When I was young (in the very old days) to find anything out I used to have to read a book. Since books were an expensive item I used to have a list of books I wanted for my birthday or Christmas and I used to visit the local library every week to sit and read and learn new fantastic things. I would get lost in a world of fantasy and knowledge; even then I had a thirst for learning which fortunately has never left me. It is such a shame that local libraries all over the world are shutting due to technology. (Do a Google search for News and Libraries and Closing)

Now information is available at the click of a button. We can find information on how an airplane works or the theory of relativity in an instance. Technology has made all this information easier to get, however IMO it has made us think less.

Do we still question all this information?

As testers we know we should be questioning everything, we learn to sort the chaff from the wheat as the title of this blog post implies. (Link). With such a wealth of information that is so easy to gather the skill is to be able to collate this information and remove the distractions. How easy as testers do we find this? I find it a natural thing I appear to do without thinking until I started to write this blog article. My concern is that with so much information do I end up throwing away something that is later proves to be vital or important.

Does anyone out there in the testing community have a method they use to help with this?

How do we not forget everything?

This then leads on to the topic of self learning – how do you select what to read and what not to read? How do you ensure you do not miss a really important article that has been blogged?

One approach I use for self learning is to use twitter and the software testing club (Link) within these communities’ people we talk about blogs they have read or recommend to read, the power of the crowd. This helps to reduce the amount of information I have to process. Another approach is to actually talk to people; humans are wired to be better at absorbing information via speech than from the written word, it is more likely to be remembered.

My other concern with all this information is our ability to remain focused, another important testing skill. With so much information to digest it is so easy to get into the habit of just scanning the information and not reading the whole article (I wonder how many people will get to this part of the article?). It is so easy to just start an article and get distracted by some other piece of information and not return to the original article. I sometimes think I should not add any hyperlinks to my articles and just add them to the end but I want to credit the people who inspire me or provide me with information as I write the article, it is one of those things which is important to me. A perfect example of this was the recently article Michael Bolton wrote about estimation (Link) which was a five part article, how many people read the whole of the article? It is so easy to skip or scan and miss an important point within an article and the same can be applied to testing. If we scan and miss something it could be that the thing we missed will cost us a lot of money.

My other concern is that we are becoming a society of 24x7 learners, we never switch off.

Are you one of these people?

My concern on this came from a conversation in which I stated I do hobby as a job, this scared me. I have a passion for testing and learning but am I not in danger of burning myself out or forgetting valuable knowledge unless I switch off?

How many others reading this blog switch off and pursue other interests outside of technology?

If you take nothing from this article please do switch off. I have hobbies that have nothing to do with computers. I enjoy being creative and take photographs, spending time outside at stupid o’clock catching sunsets and sunrises and landscapes. I enjoy growing things and spend time in my garden. I am fortunate to have a very large garden in which to grow and nurture things. I also have a family and I am a grand parent and spending time with my granddaughter is such a wonderful thing to do. We call her our little time waster – since time can go so quickly when you are engaged in playing. After all these hobbies it is surprising that I have time to do my job or to keep learning but I come back to my work more energized and ready to learn more.

Since starting to write this article I have found a couple of other blogs that mention the problems of attention span and remaining focused they can be found here:


I do recommend people reading them.

PS I will leave it to your imagination who I think the Morlocks are :o)

Friday, 3 December 2010

The Human Element

I attended the Eurostar Testing Conference (http://www.eurostarconferences.com/conferences/2010/) in Copenhagen, Denmark this year and met a large group of very interesting people. A few highlights for me were:

Meeting the Cartoon tester (http://cartoontester.blogspot.com/) in person, a friendly unassuming guy with a quick sense of humour.

The other highlight was amount of ‘real life’ examples of exploratory testing and session based testing management. One of the best things I took away was from Carsten Feilberg’s talk on Session-Based Testing in Practice (http://carstenfeilberg.blogspot.com/) in which he reframed the wording of SBTM to Managing Testing Based upon Sessions. It was a why did we not think of that before!!!!

One of the keynotes was by Stewart Reid on “When Passion Obscures The Facts: The Case for Evidence-Based Testing” in which he looked at what testing could learn from Evidence Based Medicine (http://en.wikipedia.org/wiki/Evidence-based_medicine) . During the presentation I thought I could see many flaws in the argument he was trying to put together but could not quite work out what it was. One thing I have found out since and one point that Stewart did appear to miss was the work of the GRADE Working Group which is a newer system (and appearing to gain ground). The principles here are based upon Extrapolations (http://en.wikipedia.org/wiki/Extrapolation).

To quote from Wikipedia:

Extrapolations" are where data is used in a situation which has potentially clinically important differences than the original study situation. Thus, the quality of evidence to support a clinical decision is a combination of the quality of research data and the clinical 'directness' of the data.

Interestingly the data gather for extrapolation are more based upon human experience rather than just a set of numbers. Is it just me or is this like running a set of known tests then exploring afterwards? See my previous post on Hybrid testing (http://steveo1967.blogspot.com/2010/09/hybrid-testing.html)

So why have I called the title of this blog post “The Human Element”?
I was having a conversation with my wife(Tracy) after the conference since she is a retired theatre nurse and understand the medical arena very well and she came up with a wonderful phrase. It is all well and good having all these numbers and statistics but you cannot ignore the human element. She gave an example of this in which a nurse working in Intensive Care has a lot of machinery (with installed software) at her disposal however none of this equipment can tell her if the patient is feeling happy or sad or is uncomfortable.

Tracy said the problem is no machines have a soul they do not care how the patient is feeling, the machine could be saying everything is ok but the nurse and their compassion knows and understands how the patient is. I asked my wife to have a talk with Stewart and some other testers including Lynn Mckee (http://www.qualityperspectives.ca/)

This provided a wonderful insight to me in that we as testers forget that there are lots of people who we should be using as oracles for when we test a system we should not be forgetting about the human element.

During her conversation Stewart started to mention the use of statistics as evidence and for making healthcare decisions (Cochrane Library - http://www.thecochranelibrary.com/view/0/index.html?s_cid=citation) and Tracy said that to get to the point of making a decision still requires the GP to ask questions and to explore all possibilities. At the end of the day it is just statistics said Tracy and it does not help in a situation in which a perfectly healthy 20 year old is prescribed a drug for a problem and then dies due to an undetected heart problem. No amount of journals, evidence can account for this, since it is on a personal level between the patient and the medical expert.

The final conversation I remember Tracy having was with Lynn and a few other and it is very useful for testers when they come up against the problem of ‘It should do this.’
Tracy talked about Dr Spock and the book about the development of children (http://en.wikipedia.org/wiki/Benjamin_Spock) and that at certain age’s children should be doing this and that. This book causes major worries in parents when their child does not meet the timescales within the book for talking, sitting up, walking etc. Tracy then made a point which caused a great amount of laughter. “People seem to forget that babies have not read the book – they will develop at their own pace”

I found this a wonderful piece of insight, we seem to forget that everyone is different and if we apply this to software and the development of software we start to realise that every piece of software is different and that we need to explore the software and play with it to get the full potential out of the software.

To conclude this post I would like to say a big thank you to my wife Tracy for her encouragement and support in what I do and for giving us testers a lesson in remembering about the human element in what we do.