Monday, March 16, 2009
Testing the Truth
This got me thinking: would I have an ethical objection to testing a project like this? If not, what kind of project would I be unwilling to participate in? What if I were in a project and suddenly realized that the product could be used for nefarious purposes? What if my superiors responded to my protestation with an emphatic, "meh...get back to work."
So here are some questions I ask the greater testing community:
(1) What role do your ethics play in your software testing?
(2) Is there a project you would not work on, or have not worked on, based on moral objections?
(3) Does the current economic climate change what you would/would not work on?
(4) How would you test a product that claims to discern the truth?
(5) Would you purchase an iPhone app that could discern truth from lies?
Friday, March 6, 2009
Learning to SHUT UP!
There will be a day in your career when you will implement something you've always wanted. It may be a tool. It may be a process. It may be both. But it will seem to you as though it is yours.
You will nurse your burgeoning idea from cradle, through its shaky infancy, all the way up to the point when it begins to walk "on its own". You will grow more proud each day at your child's accomplishments. How it tidies up after itself. How polite it is to newcomers. How well-behaved and dependable it seems to be. You will swell up with pride as you show passers-by pictures of your "idea" doing the most wonderful things - running, spawning, making reports. And though you will fiddle with the thing for what seems like an eternity, you know that you know this thing, this idea, this slice of heaven - born from the marriage of your intellect and your soul - will undoubtedly change the course of history for all who would gaze upon its perfection!
If only everyone else would look at it, they would see it as you see it.
And then someone will call. Someone always calls. There is a problem. And this time the problem is not going to be solved by your miniature doppelganger. Actually, this problem was a direct result OF YOUR IDEA! Perhaps it was an oversight you made while extolling the virtues of your chosen child. Some inattentional blindness you succumbed to as you wrestled with either analysis paralysis or were being beguiled by some new technique you wanted for your resume. The once crowning, glorious edifice to all your rational logic is rent asunder by the astute observation of those you placed in subjection to it.
Your first emotion will be one of incredulity. Certainly the problem is elsewhere in another system...one you're not really responsible for. But no, the evidence points to your own backyard. Then you will seek to affix the blame on someone else. Someone else should have caught this! But alas, you'll be unable to prove that empirically and you'll realize that any effort to do so will only cause an escalation to occur. You'll be trapped. The problem finder has been found to be...a problem!
Oh, the insult to your abiding genius!
Now listen to me very closely. During all this frenetic activity, you will be tempted to lash out at those who accuse you. And you'll know all the words to say to give you the appearance of rightness. A quote here. A name-drop here. A book you read over there. Or even better, you know you could go trolling for evidence in your data farms and find the EXACT MOMENT when that ungrateful so-and-so really dropped the ball. Sure, it'll take an hour of research, but you KNOW its there. You'll even start up Agent Ransack. Are you still listening? Then hear this:
Stop it. Stop it NOW.
What you are about to do is introduce an unnecessary complexity into an already complex situation. That complexity is your personal desire to be seen as right...all the time. It is not only unnecessary, it is counter-productive.
I sometimes detect a hidden thread in our software testing conversations that bemoans our meager status. "If only they would listen to me," it seems to say. Much talk is given to methods and practices we might employ that would bolster our believability index. And this is good, but without restraint, maturity, and wisdom it leads to a kind of methodological mutiny. You can end up seeking to change the course of the development effort in a subversive manner; team buy-in be damned! This is not being progressive. This is not being helpful. This is not being clever and it will not land you a job at <insert mega-trendy forward-thinking collective here>.
You work in a team. And a team works together. You perpetuate the us-versus-them mentality at your own peril. You will make mistakes just like everyone else. Own them. Working with others is a messy affair, mostly because we are all so darned complex. We defy any attempts to metricize our behaviors, but that does not stop us from trying. Future history will be littered with our grandiose ideas to reduce complexity and manage expectation. Some may actually work, but they will only work when they work well with others.
On the cabinet above my desk, there is a post-it note stuck to the door. It simply reads:
"This is not the kingdom you are building."
Thursday, March 5, 2009
Learning to lead...
In keeping with the "script", we addressed the issues noted in their respective reviews. We talked about career goals and such. And having done this, I confirmed that they were satisfied with all that was written and addressed up to that point.
"Yes," they would dutifully reply; "Check," my mental reply.
But then I expressed to these amazing individuals that I felt responsible for equipping them not only for their jobs, but also the greater market. I spoke candidly and frankly that because we live in such uncertain times ( haven't we always? ), I felt morally obligated to assist them in becoming all I know they can be. And I apologized for failing them so in the previous year; for my distance, my myopic nature, my personal ambition making me unavailable at times.
I gave them the power and right to hold me accountable in my others-improvement endeavor. They can improve me as I help improve them. I want so desperately for my direct reports to succeed. It causes me much inner-turmoil to think that should a pink slip appear, it could send them out into the current job market with limited exposure to advanced topics in development, testing, and automation. This is my responsibility, so I endeavor to rectify it.
Managers and Testers reading this: what are YOU doing to improve the quality of the PEOPLE around YOU in ways that will benefit THEM should they no longer be around YOU?
Humanity, Suffering, and Software
Father Tutu was questioned about the motivations behind forgiveness in the face of unspeakable humanity. His answer involved describing the inherent intertwining of our humanity, and that any practice that dehumanized an other will have a reciprocal affect on the dehumanizing agent. It was incredible to watch someone who spoke with passion, precision, and humility about the betterment of life through love and forgiveness; that aspiring to the better parts of ourselves is a noble endeavor. He left the usually talkative Craig speechless. He is a peacemaker in every sense of the word.
And after this, I read James Bach's blog entry on how quality is dead. Long live quality! I will not attempt to summarize, but James' thoughts are lucid and illuminating as usual. Please read it. Now, James would be the first to say that he can be argumentative. If he is a peacemaker, it may come about only after much heated debate.
Sometimes when my mind encounters a striking dissimilarity between two things, I seek to find their commonality. What does software quality have to do with Apartheid? And so it is late enough at night for me to entertain the relationship between these two seemingly disparate media events and their perpetrators. To explore the connection between Father Tutu's cause and James Bach's cause.
Perhaps James' observations speak to a very small kind of oppression forced upon us by a culture intoxicated by its technology (in-techno-cated?). Perhaps Father Tutu's response to a particularly nasty, high severity problem ( i.e., Evil ), and his subsequent success, is a valuable model to combat the present state of affairs.
Both ideas seem to be crying out for change that benefits everyone, but are dealing with entities reticent to change because, well, there's no incentive to change; reduced quality does not hurt the 'haves'. And these entities fervently believe that nothing is wrong, that things are actually improving.
It seems to me that restoring a proper perspective is a shared idea in both issues. It is good to see ourselves as humans again - linked to each other in inexplicable ways that technology and government cannot improve upon. To begin thinking of what the word "enough" really means. In both contexts. Bringing awareness to the greater community begins to break the cycle. It all reminds me of something my Pastor father told me one time: "Some demons are killed immediately, others are starved to death." Change comes, only the velocity varies.
I'll end my philosophical waxing here because I think it is a bit disrespectful for me to equate the plight of human suffering in Darfur/South Africa with proprietary protocol failures between Tivo and Dish Network.
Monday, February 2, 2009
Google attributes bug to human error...as opposed to what?
Wednesday, March 5, 2008
Spinning Plates
"Around release time, I use the analogy of spinning plates. I like that analogy for several reasons.
1 - the activity of spinning plates requires constant intervention
2 - adding a spinning plate is easy; maintaining it amongst all the others is difficult
3 - spinning plates have an implied order, but can fail out-of-order just as easily
4 - those involved in spinning plates cannot rest...EVER.
5 - you only notice spinning plates when they stop spinning (eg, they fall off and break )
6 - to be successful at plate spinning, you have to move MORE than the plates do
7 - once you start, it is not easy to stop spinning plates without breaking something
To me, this perfectly describes the current process of delivering updates and releases; why success is so hard and failure is so easy."
Monday, December 17, 2007
The nurse who said, "You're going to die."
Elisabeth Hendrickson posted about a topic of conversation at this year's LEWT. It dealt with an item I'm very tempted by: problem diagnosis and debugging by the tester. My comment was appropriately passionate but may have been too lengthy for a polite comment. I've taken it and placed it here with only minor modifications.
One time, I went to the doctor. That is amazing in and of itself but is not the point of my comment.
I had gone to have a mole looked at. The nurse, who was busy running all the setup routines in anticipation of the doctor, quipped that it probably was cancerous. “Most people develop some kind of cellular anomaly in their lives”, she said with all the cold and callous concern a butcher might have for some hamburger meat.
Her “diagnosis” did not set well with me. Part of my proficiency for testing is my predilection for pessimism. It caused all sorts of undue stress and anxiety. My mind ruminated on all the tearful activities that awaited me; saying goodbye to friends and family, what music will they play at the service, etc. Mind you, the doctor hadn’t even SHOWN UP yet!!! When he did finally show up, he allayed my fears with a more educated diagnosis - “you’re just mole-y. Why are you crying?”
That nurse by virtue of her position/location brought to bear the air of authority. She certainly said all the educated words. But given the circumstances, all that my mind heard was, “you’re gonna die.” How was I to know?
I love to break things so I can see how they work. And when something breaks, I view it as an opportunity for me to observe the thing’s inner-workings. But I’ve been able to avoid freaking customers out by not doing such diagnosis in front of them. I save that conversation for the developers lest I come off as an authority. Or much worse - the nurse who said, “you’re gonna die”.
