Category Archives: Lean

Q&A A Fistful of Spreadsheets

[Note from the author: after some fact checking I discovered that I used the Red Card in a wrong way in this blog post. The red card must be used ro raise issue like low volume or high temparature in the room, which can lower the quality of the gathering of information. For normal interruptions the yellow card should be used. This basically menas that I have to rewwrite the blog post on certain points.]

Facilitator: Welcome back to another round of questions and answers. This Q&A will focus on the serie Smart Use of Spreadsheets of Mindful Tester.   K-cards will be used in order to structure the flow of thoughts.

I’ve got green cards from numbers 43, 25, and 9. Number 43, you can ask your question.

Attendee number 43: I only read instructions. Why do you not add more clarification?

Writer: It is important to discover things myself. This way I get a better understanding of spreadsheets. In this serie I wanted to share this experience with you, the reader.

 

Facilitator: I’ve got yellow cards from numbers 31 and 16. Number 31 it’s your turn.

Attendee number 31: That’s pretty abstract. I don’t get that. Would you provide an example?

Writer: In the last episode I described the use of Remove duplicates in order to get a set of pizza brands or a list of names, which are mentioned only once. During the test I needed a way to make a set. I explored the spreadsheet program to find an easy way. By letting the reader follow my steps she or he might see some functions of spreadsheets, which are unknown and useful in the future.

 

Facilitator: Number 16.

Attendee number 16: How did you find it?

Writer: By selecting a tab, which might contain an interesting set of functions.

Attendee number 16: Just curious. Did you know the function before the test?

Writer: No.

Attendee number 16: So you stumbled upon it during the test?

Writer: Yes, I did.

Attendee number 16: Is it not strange to start testing without an exact set of instructions?

Writer: No. Let me elaborate.

Attendee number 16: Sure.
[Facilitator relaxes.]

Writer: Is it possible to make a test strategy, which does not change? I think: “Not”, so making test cases should be postponed as long as possible or skipped when possible. Nothing is worse than hours spent for nothing. The same is also applicable for the use of tools to determine the most easy way to test the system under test.

 

Facilitator: A yellow card from number 33. Number 33 it’s your turn.

Attendee number 33: When do you skip test cases?

Writer: There are other alternatives like checklists and exploratory testing.

Facilitator: this is going off topic, so I am closing this thread.

 

I’ve got green cards from numbers 25 and 9. Number 25, you can ask your question.

Attendee number 25: In your blog post you are using a spreadsheet program. Why did you not use a database tool or Python? They are better equipped to handle data.

Facilitator: I see a red card from number 493. You have a question about the question?

[Attendee number 493 nods.]

Facilitator: Number 493.

Attendee number 493: what is Python?
[Facilitator looks for a snake.]

Writer: For me is Python a strong light weight programming language. Do you need more information?
[Attendee number 493 shakes his head. Facilitator relaxes.]

Writer: The question was: “Why did you not use a database tool or Python?”I have to take additional steps: install tools and figure out, how to use them. It would have taken me extra time.

 

Facilitator: I’ve got yellow cards from numbers 1 and 24. Number 1 it’s your turn.

Attendee number 1: This could have been a nice starting point to explore another tool?

Writer: Yes, but… The but is to look for long term benefits. For simple data manipulation spreadsheets are good enough. I did not expect complex data analysis in the near future, so I did not even consider using other tools.

 

Facilitator: Number 24

Attendee number 24: why did you not use the experience report to tell your story? This would provide more insight in the way you work.

Writer: it is nonlinear and it would confuse the reader, who is looking for a fast and linear solution.

Attendee number 24: Could you be more specific? I mean: could you share an experience?

Writer: The first time I used Remove duplicates I got the wrong combination of pizza brands and phone numbers. So I did a manual check for the remaining wrong numbers. Later I discovered that I ordered the phone numbers from A to Z instead of Z to A. So I used System 1.

Facilitator: I see a red card from number 493. Number 493.

Attendee number 493: what is System 1?

Writer: System 1 is used for automatic and fast decisions like walking to the supermarket. System 2 is used for the complex operations, which need thoughtful thinking like buying items on the shopping list.

Attendee number 24: So?

Writer: When I heard the word Order, I picked the order from A to Z. For System 1 this is the default choice. I had to use System 2 (Is the order really from A to Z?), before continuing the test. Adding this experience is likely to add confusion for the reader.

Attendee number 24: I think that you are smart enough to make a smart formula to automate the manual check.

Writer: Thank you for your compliment. There were only foursome numbers to check. I actually made an attempt to use a formula, but that costed me too much time.

 

Facilitator: Are there any more questions on this thread? Apparently not.

I’ve got green cards from numbers 9, 10, and 11. Number 9, you can ask your question.

Attendee number 9: Why did you not use a programming language supported by the spreadsheet program?

Writer: I didn’t know the language.

 

Facilitator: Number 10.

Attendee number 10: Why did you not use a blink test to compare?

Facilitator: I see another red card from number 493. Number 493.

Attendee number 493: what is a blink test?

Writer: A blink test is a test, during which the tester processes a lot of information in incredible short time or in a blink of an eye. In this particular case you can switch fast between the two spreadsheets. Differences can be noticed: characters change.
[Attendee number 493 nods.]

Writer: The blink test has some disadvantages. Spreadsheets can be so big, that you need a lot manoeuvring. Some spreadsheets have cells with default width, so not the complete content is shown to the tester. Non-readable characters can be contained in cells. My favourite is that trailing spaces cannot be spotted with the naked eye. “Text” looks the same as “Text<space><space>”.

 

Facilitator: Number 11. You can go ahead.

Attendee number 11: You copy the calculated values instead of the formulas to the last worksheet. [Speaker nods.] That is a smart move, but it looks like a redundant step to me.

Writer: if everything is OK, then you are right. It there is more than 1 NOK, then odds might change for me. In this case I have to verify the formulas. If they are right, then I have to analyse the data. Values can better be manipulated than formulas.

Attendee number 11: Like?

Writer: Spreadsheet programs support auto filter, which contains nice features.

 

Facilitator: I’ve got green cards from numbers 21, 562, and 23. Number 21, you can ask your question.

Attendee number 21: You already advocated against the use of database tools and programming language in this context. Why did you not use spreadsheet comparison tools?

Writer: It would only save me one step: to learn the tool. I have to admit that I am not familiar with these tools.

 

Facilitator: I’ve got green cards from numbers 562 and 23. Number 562.

Attendee number 562: Why do you not change the background colour of the cell to red, if the cell contains a NOK?

Writer: It is an extra step. For analysis of complex patterns of NOKs it can be considered.

 

Facilitator: Number 23.

Attendee number 23: I found a neat trick. If I have to compare a new sheet with one in the map, then I copy the sheet in the map and rename the tab. What about that?

Writer: It is a tricky one: after months it is not clear, which spreadsheets have been compared. More particular: what were the sources of the sheets? Also there is a risk of different numbers of rows. You actually have to refactor the spreadsheet.

 

Facilitator: A yellow card from number 21.

Attendee number 21: Could you give an example of refactoring?

Writer: I would an extra row above the table and add a text with information over the source. You still have to watch the formulas though.

Facilitator: A green cards from number 372. Number 372.

Attendee number 372: Why do you not use the standard compare function of the spreadsheet programs?

Writer: Good question. I’ll have a look.

Attendee number 372: Sounds like a new blog post?

Writer: The last blog post about spreadsheets took a considerable amount of time.

Attendee number 372: Why?

Writer: I was refactoring the solution.

Attendee number 372: Is that not contradictory with your previous statements about being lean?

Writer: You’re a smart girl. I blog in my free time.

 

Facilitator [a bit surprised]: I see a green card from number 493. Number 493.

Attendee number 493: What do you mean with “=b”?
Writer: “=” is an underarm and “b” a hand.
Attendee number 493: Thanks. =b.
Writer: Thanks.

Facilitator: Thank you for all questions about spreadsheets, testing, and… Hey, I miss the pizzas. Anyways I hope you enjoyed this Q&A.

Keep on questioning things!

 

Fetching, fast and slow

Let’s start with a simple observation of the following sentence:

Do I communicate (without a mouse)?

 

If you were fast, then there is a high probability you were thinking: “The word mouse is used by the author instead of the word mouth. So the spelling is wrong.”

If you thought hard, then there is a high probability you were thinking: “The mouse is probably a computer accessory. So the question could be rephrased as follows: do I communicate (without a computer)?”

I once heard a great story from an experienced test manager René. He told me, that his project members were communicating with emails. It did not even change, when they sat in the same room. He just set a daily limit of 3 emails, which they could send. This lead to more face to face communication, which improved the project spirit and group cohesion significantly.

Breakdown
In the introduction I used a mind trick on you. This is a trick, which can be used to confuse people.  For the fast observations System 1 is used in most cases. This way of thinking provides fast, almost effortless way to digest information. Like walking to the office.

For the thoughtful observations System 2 is used. Doing complex operations like testing in the office. Both systems have been discussed in depth by Daniel Kahneman in the book Thinking, fast and slow. The title of this blog is based on the title of this book. And fetching is short for fetching coffee.

In March 2015 James Bach introduced the term testopsy. He analysed, what the tester did during a test. I thought about a post autopsy or blog post autopsy.  It is tempting to concatenate the strings post and opsy. But as a Dutch native speaker I do not take any chances. 

Just let me perform an autopsy for a blog post: how did I construct the blog post Do I communicate (without a mouse )?

Because I wrote the blog post, I have the original mind map, which was used as a basis. In the following picture I highlight, which System is used and the corresponding trends. System 2 is used in certain parts of the blog post and System 1 in other parts.

mindful-tester-mindmap-System1-Sytem2

A manager would say:
“This is nice. And that’s all. The relationship between the parts of the blog post and Systems is meager at most. ”

Time for a graph make over.

Graph SOS
There is a British car program, in which they deconstruct and construct a beloved car wreck. For people, who earn it.  And as a reader you definitely earn a better graph.

In case you did not read this blog post Do I communicate (without a mouse)?, please do. It makes the following graph understandable.

 

mindful-tester-timeline-system1-system2

In the graph above I show my usage of Systems 1 and 2 on the vertical axis. The horizontal axis shows the different parts of the post in reading order. So it is possible to observe, that System 2 is used less and less until the end of story.

A manager might be more interested in this graph than the previous one:
“So what you are basically stating, is, that people use System 2 for learning. And participating.”
Or even better:
“I wonder, whether System 1 is used during the meetings in my company .”

Breakdown 2
Over the years I saw a recurring pattern. Every time I boarded a project or got a new room, I had to change my coffee fetching list. And somehow I reduced the spent time. I eliminated waste: it looked to me, that I was lean. This story I carried with me for more than a year. Telling and retelling it to myself over and over again.

Then it was time to put it in a blog post. I started with the mind map with the condensed and descriptive title Coffee. The first branch contained the story. Then I added two funny anecdotes to add some flavour to the blog post.

In the meantime I had lost my favourite mug out of view: a bear, who juggles, while praising the owner of the mug. (That’s me.) After I had found the mug, I made a photo with my smartphone. The mug was on the foreground and my markers right behind them. Then I noticed my mouse: it was behind the markers and not on the photo. So I changed the view for the next picture: the mouse is on the background. Unreachable for normal use.

A few days later I noticed, that I was missing a photo with a funny text for the blog post. I needed something, that could be connected with coffee or tea. Then I remembered the picture of my mug, which was a major obstruction for using my mouse. A thought about communication entered my mind: Do I always need a mouse for communication? It was relatively simple to write an introduction from this point.

Most stories have a lesson at the end. I think, that it is highly predictable (and a bit boring). I wanted to give the reader a choice out of 3 lessons. But that was not entertaining enough. So I placed myself in the spotlights (again). If I could let my voices speak, then I would have a more recognisable situations instead of some abstract and concise questions without any explanation. I took the following voices:

  • The lean machine in me, cutting wastes on his way to the future
  • A woman constantly looking for her needs, while brainstorming and chatting
  • A service desk agent concerned about an implementation of a new functionality
  • A curious software tester looking for clues.

I somehow used a Dutch style form: the circle is round. I started with the title Do I communicate (without a mouse)? And ended with the same question.

Do I communicate (without a mouse)?

Just a bunch of thoughts popping up in my head, while looking at the picture:

  • Yes, I need some tea.
  • Yes, I need to make a note.
  • Yes, I need a mouse to communicate.

A Note As A Service
In the Netherlands people tend to be too busy to go to the coffee corner in the office. In some companies there is an unwritten rule, that you get coffee, tea, or water for your colleagues once every 4 hour.

The first times are hard. What is your name? What would you like to drink? How do you prefer to drink your coffee? Coffee is a difficult one. How strong, how much milk, what kind of coffee ? If you have 10 waiting colleagues, then you need some time to note their requests.

Because I am Chinese, I can make the following joke to make people relax. “So you take numbers 7, 14, and 22.”
Then  I get surprised looks.
Probably thinking: “Our coffee machine has no number 7.”
Then I continue with a heavy Chinese accent:
“One Babi Pangang, one Fu Yong Hai and …”
Then often a smile appears.

Accelerated note taking
Of course this process can become faster than I described. The names of the persons were abbreviated by me. I used codes for the beverages like C strong for a strong black coffee.  The waiting time during a phone call could be reduced to seconds by asking his or her colleagues: “What does he / she prefer to drink?”

I once noticed a serving tray. The type, which is used in the canteen. There were more than 20 circles on it. Every circle contained the name of a person, preferred beverage in the morning, and preferred beverage in the afternoon. When I called an end user with question about domain knowledge, I got the reply, that she was fetching coffee for her colleagues. So I told, that I would call back after a half hour.

Association and reduction
The next trick was to discover patterns: he always drinks black coffee in the morning. Or she prefers hot water for her tea.

Visualisation is also great: imagine the face of someone you fetched coffee for: cappuccino. Or look at the desks and the corresponding beverages: this is the tea corner.

My question became: “Would you like to have a black coffee?”
A few weeks later: “Black coffee?”
A few weeks later drinking an imaginary cup of coffee and waiting for a nod.
A few weeks holding an imaginary cup of coffee and waiting for a nod.
A few weeks a slight raise of the chin and waiting for a nod.

Questions I ponder upon
(
To spice things up I added some fictional thoughts and talks. )

  • Am I lean?
    I fetched coffee and tea for 8 persons within 8 minutes.
    Not bad!
  • Can I handle changes?
    “For a change I would like to have real hot water from a water cooker. Earl grey, sugar, and a real tea spoon. I hate those flimsy plastic reeds. Can you still remember it? The next time I’ll fetch you some tea. Or maybe I should take green tea. My neighbour really loves it. Nah, I just stick to the dark tea. Anyways….”
  • Do I communicate (without a mouse)?
    “Yes, Earl Grey. By the way I noticed, that you are testing the upload function. There are customers begging for it. Did you know that?”
    “Really?”
    “I was wondering, whether the following item is mentioned in the user story!?”

 

How to Plan a Visit To Amsterdam

Board games are a great way to pass time with children. One of my favourites is Amsterdam. In this game you win, if you are the first one, who have visited 6 random chosen touristic places in the Dutch capital.

Planning as usual

At the beginning of the game my children and I had each 6 cards with touristic places to visit in Amsterdam. So I told them to find the places on the map. Then they had to plan their visit using the canals with a boat. This was more difficult. With some help from me they planned a boat ride by making a stack of 6 cards.

Replanning as usual

In the next step each player placed two bridge blocks, which can be compared with road blocks for cars. Six bridge blocks had a great impact on the planned tours. The ideal routes were blocked or even worse, touristic places were blocked. So the whole process of stacking cards began again. My children did not complain. 

Planning in progress

During the game the bridge blocks can be moved by players. So carefully crafted routes became obsolete. This time I was smart: I adviced my kids to plan a route within a small area of their boats.

Heuristic as a planning

After visiting a touristic place the player gets a yellow card. This can be a nice event like moving bridge blocks or distressing one like going to a boat of another player. My planning advice changed to: go to the touristic place, which can be reached by passing a minimum of bridges.

Planning as a service

So long planning saves time and reduces frustration, it is fine with me.

 

Am I lean, doctor?

This summer I went to an extraordinary museum about cars. In the Louwman Museum in The Hague cars, which are landmarks in the history of the automobiles, were shown. The range was from motorized carriages to more familiar cars on the road. The owner had also a special interest in strange cars. I discovered, that electric and hybrid cars were already used at the beginning of the 20th century. Then a piece of furniture drew my attention: the desk of Dr. Toyoda. He worked for Toyota, which has a special approach for Lean Management.

Plans 2.0

For the test project I had two iterations, which took 3 weeks each. My planning for an iteration was simple: two weeks for the functional tests and the last week for the user acceptance test. To be more precisely 40 hours in the last week for 3 end users. I mailed the test plan to everyone and I got polite decline from the end users. In a normal week they had each 8 to 12 hours left in a week. So ideally 36 hours of testing would be filled in. But it was still too short. Overwork was no option. It was time to reschedule the activities.

A few months before I had bought The Toyota Way. It was mentioned several times by experienced Lean practioners, so I bought after browsing it. Heijunka, Level out the work load, came to my mind. If I somehow could move the testing hours of the end users towards the beginning of the test iteration, then the planning problem would be solved. Suppose function A was successfully tested during the Functional Acceptance Test (FAT), why should I wait to let this function tested by the end users? If FAT went well, then the first functions could be tested by the end users in the first week of the iteration. So I made new calculations: in worst case I have 3 end users, who have 8 hours left in 3 weeks, then I come to a total of 72 hours of testing. The new test planning was easily accepted by the end users.