Adam White

I was impressed by what Cong had done. I thought his summary was thoughtful and showed understanding of the concepts. I went through bugs and reviewed the answers to each of the 4 questions. I learned a few things about the details of our product that I didn’t know before.

I picked up the phone and called him to talk to him about his findings. I gave him credit for doing what he did. When discussing the bugs I asked him if he could do this process for every bug or if it would be too time consuming. His answer was something to this effect “Well even for bugs that I thought were trivial I still learned something from talking with the developer so I’ll have to use judgement in the future”

In discussion of one particular bug I asked him to think about whether it was something we should write down as a test idea for future release and also to help future testers learn about the product. I didn’t give explicit directions/instructions on whether this is an expctation/requirement for him. He will have to figure that out for himself and decide where the info would go.

One point we discussed was that there was no response to my initial request until it was done. Throughout the week I wasn’t sure if he was putting any effort into it. I’m also at fault for not setting expectations for an initial response. We both learned a lesson from this and can now communicate more effectively in the future.

I feel like now when I respond to his daily updates we can speak a common language and have a common understanding.

Here is my summary of this interaction

What went well

• Attempting to make Cong feel safe and that I wasn’t threatening or unhappy with his work. I am his boss’ boss afterall and I had never given him a direct request before
• Clearly set completion date
• Set out expectations on how the end result should be delivered
• Gave some local options for getting help
• Follow up via phone to discuss findings and do a little bit more coaching

What could be improved

• I could have picked up the phone in the beginning and explained the exercise, following up with the email writeup
• In my initial request I could have asked Cong to respond and either accept, reject or negotiate my request and give him a date to do this by
• I could have followed up with him mid-week to see how things were going with the task
• I could have set up a skype call with him

Things to remember when coaching remotely

1. Look for things that will teach you about something of interest to you – not only the person who you are coaching
2. It’s important to set two deadlines – one for the initial response/commitment and one for the actual deadline
3. If you’ve not met the person or had very little interaction with them make sure they don’t feel threatened
4. Give the person options for getting local help
5. Follow up via phone/skype, give feedback watching for anything from #1

I’d love to hear any comments on this situation or thoughts/experiences you have with remote coaching/training

Cong’s response  (only edited formatting)

<begin response>

Summary

One of the most important things in testing, is to have a sufficient number of perspectives. We can either look at a product by its elements: Structure, Function, Data, Platform and Operations-SFDPO, or by quality criteria from customer’s point of view, CRUSPIC-Capability, Reliability, Usability, Scalability, Performance, Installability, and Compatiblity. Note that scalability and performance are closely related, and usability, what might annoy or frustrate the person or program that uses the product, and installability should not be forgotten.

In test design, risk-based testing will try to anticipate a problem and test for it. Rapid Testers model the risk in four parts. There is a risk when a victim may be affected by some problems caused by a vulnerability in the product that is triggered by some threat. The model itself is a good application to the exploratory testing we do here at Platespin. When deciding how and whether to test the risks on the risk list, where risks are assessed via a combination of the four parts, tasks are prioritized and used to guide exploratory testing.

As a tester, especially a tester in exploratory testing, we examine the criteria (CRUSPIC) from a user’s point of view. Surprised, annoyed, puzzled, frustrated, or maybe satisfied, emotions are generated when user experiencing our product. Emotions, “arousals”, should guide us in testing and work as oracles that trigger heuristics and wake us up to the discovery of a new problem, a problem that might be function-wise correct but interfere customer’s goals.

In addition, there are several key points that can lead to successful testing: A diversity of workflows tends to expose sequence-related bugs; A scriptable or programmable API to a product makes it more testible (which is the case that we use vmware ESX and vCenter) ; shortening the feedback loop between developer and end-user is a powerful force for beneficial change; real (realistic) circumstances tends to reveal problems and to help identify the kinds of problems that are important to real users.

5 Bugs discussed with developers

Edit – removed bugs that were discussed

Please point out anything you like.

Thanks,
Cong
<end response>

I thought I would share something I did recently regarding remote coaching/training. My inspiration for sharing comes from here and here

I have always enjoyed training new software testers. Lately it’s been more difficult for me to do since I’m in a different office from the bulk of the testers on my team. See my linked in profile for an explanation as to why I’m away.

Given that I’m away I started looking for ways to continue engaging with my team. One opportunity came from reading Cong Zhao’s daily updates.

About Cong: Cong joined us from Waterloo for his second summer work term. His first work term was with ACRP. They provide testing fixtures to Rim to test blackberry. You can find out more about him on linked in

About daily updates: I have everyone on my team send a little summary of the interesting things they learned that day. It’s meant to resemble newspaper headlines that catch your eye, make you want to read further and hopefully learn something of value to you.

Note: I’ve asked Cong for permission to share the process we went through.

Here is the initial correspondence I sent to Cong based on something I read in his daily update.

Cong,

Mike and Herb are out so I’m going to do some remote mentoring with you. I reviewed your list of bugs. You did a good job at re-running the original steps and documenting that in the bug. I want to challenge you to figure out what other areas of the product might have been affected by the changes you are testing.

I want you to ask yourself (and others) “What else could have been broken as a result of this fix?” “What areas of the code where changed and what other areas depend on/use this area” This will cause you to become familiar with the code and will probably require talking with the developer who fixed it.

Questions you should be able to answer for every bug

• What area of the code was changed?
• Why was it changed (explain why the change fixes the bug).
• What needs to be done to verify that the fix resolves the original problem.
• What needs to be tested to prove that the fix does not cause the product to break in some related but different area.

For some guidance on different approaches to thinking about the software check out these articles

• More Than One Answer More Than One Question
• Is There A Problem Here
• Test Design With Risk In Mind
• The Proof Of The Pudding

Action items I expect you to complete by next Friday.

For Articles listed above

• Read these articles over the next week
• Write a summary of your learning from them
• Send summary to Mike, Herb and myself

Look for and document your exploration of bugs beyond the original description.

• Pick 5 issues that had code changes that you are assigned to close.
• Speak with the developer and get the questions from above answered. Document the answers in the bug when closing.
• Show evidence of the above to Mike, Herb and myself.

Given that Herb and Mike are out and I’m remote – you could also rely on Joseph for som
e guidance as well. Let me know if you have any questions or if you cannot complete this assignment by next Friday.

Thanks,

Adam

What I think I did well with this communication

• Explained to Cong what I was intending. I attempted to make Cong feel safe and that I wasn’t threatening or unhappy with his work. I am his boss’ boss afterall and I emailed him fairly out of the blue.
• Clearly set completion date
• Set out my expectations on how the end result of the work should be delivered
• Gave him some local options for getting help

What could have made this communication stronger

• I could have asked Cong to respond and either accept, reject or negotiate my request and give him a date to do this by

I’d love to hear any comments on this situation or thoughts/experiences you have with remote coaching/training

I was using a mind mapping tool called BrainMine. I found a bug that exhibited itself if you printed the mind map and then edited and/or moved a node afterwards. What do you think is going on underneath the covers here?  Would your automation have caught this? How would you automate this test?

The video is here

I wasn’t up on my BB Test assitant skills when this video was taken so you’ll have to infer my troubleshooting process from what you see in the video.

I mentioned that I would post more about the acquisition of PlateSpin by Novell. I’ve been delayed in doing tihs because there isn’t that much to report. We are being told that it’s business as usual and that’s what it’s turning out to be. Novell hasn’t had a strong presence in the development area – although this isn’t true for other areas like sales and professional services (from what I can see). There hasn’t been any fall-out or mass exodus of employees. 

I’ve heard people talk about acquisitions like the exodus is the “norm” either because it’s a technology acquisition and the acquirer doesn’t keep the employees or because they are scared that will happen. There have been little rumors going around about this person or that person leaving but nothing can directly link to the acquisition.

(more…)

What do I do with this information as a user? I don’t know how to troubleshoot this – nor do I have time to. Why can’t it tell me what RSS feeds it’s having a problem with.Of course I searched for the error message using google. I searched for “outlook RSS feeds error” and “RSS feeds error 8004010F” neither of which yeilded a link to a solution.

I like to do testing exercises with my team. Below is a list of links to various activities I have done with the team at some point in time. I usually turn these things into an interactive session. I thought I would share my list with you in case you want to do some of them with your team (or by yourself)

Observation skills
Which way is she spinning?

A similar picture

Cartoon
Powers of Observation

Activity – watch videos and answer questions at the end
New TV Show

Observation – General Link
http://www.smart-kit.com/scategory/brain-art/>Brain Art

Are the two blocks different colors?

How many differences can you find?
Spot the Differences

Activity for Team
How many F’s?

Activity
Optical Illusion

Cartoon
Perception

Video
The Amazing Color Changing Card trick

Where  possible with these exercises I get the team write out their observations before we go into discussions about what they see. This is to help prevent certain biases from showing up – mainly confirmation and hindsight biases

Confirmation bias, according to wikipedia, is a tendency to search for or interpret new information in a way that confirms one’s preconceptions and avoid information and interpretations which contradict prior beliefs.

Hindsight bias, sometimes called the I-knew-it-all-along effect, is the inclination to see events that have occurred as more predictable than they in fact were before they took place. Hindsight bias has been demonstrated experimentally in a variety of settings, including politics, games and medicine. In psychological experiments of hindsight bias, subjects also tend to remember their predictions of future events as having been stronger than they actually were, in those cases where those predictions turn out correct.”

When we did these exercises it made for some really interesting discussions – especiallly the spinning dancer and the different color blocks.  Doing these exercises can help you become aware of (and possibly manage) confirmation and hindsight bias.

Places you might find more information and exercises
Inattentional Blindness
Selective Attention

It’s been a while since my last post. I’ve got lots to write about and not enough time to do it.

 The one thing I do want to mention is the upcoming CAST conference which will be held in Toronto this year. Some of the best minds in software testing will be at this conference. Speakers like Michael Bolton, Robert Sabourin, Jerry Weinberg, Cem Kaner. Not to mention I’ll be doing a presentation as well

The full program is available on the website – you can find it here http://www.associationforsoftwaretesting.org/drupal/conference

Make your reservations soon – Jerry’s tutorial is selling out quickly (if it’s not already sold out)