Alas! regardless of their doom,
The little victims play;
No sense have they of ills to come,
Nor care beyond to-day.
That is part of a poem by Thomas Gray, and I’m thinking about it now as I sit in my home office, indefinitely isolated from my family. Like nearly everyone around the world, my life and my business have been derailed by the Coronavirus. (WordPress, with the precious simplicity of AI, suggests that I might have meant to write “coronations” as the final word in that sentence).
I started seriously tracking the crisis on January 27th. By that time it was already difficult to get masks, though we were able to get lots of hand sanitizer before the rest of the world caught on. By February 10th, I had learned that masks aren’t helpful for walking around in public (unless you already have symptoms). I brought one with me anyway to my gig at Texas A&M, in case I was stuck sitting next to someone with a cough. It seemed like there was a chance that the disease could be contained, until the Diamond Princess debacle. Then I realized there was nothing stopping the exponential growth of this disease, worldwide. The Chinese, with utterly draconian measures, have only slowed it down. If they return to normal life it will explode again.
It’s particularly hard to be a professional critical thinker at a time like this. My job is all about seeing the truth about trouble, and there’s altogether too much of that truth coming at us. I spend a lot of time reading research on Covid-19 and epidemiology, then explaining to my family how terrible this will get. My son and I are competing for who will be first to uncover the most depressing facts.
Yet see how all around ’em wait
The ministers of human fate
And black Misfortune’s baleful train!
Ah, show them where in ambush stand,
To seize their prey, the murderous band!
Ah, tell them they are men!
Indefinite Isolation
On my most recent trip, I canceled all dinner engagements (after reading a case study of a Hong Kong family who got infected over the course of a single meal together) and wore gloves all day while teaching. No one else was canceling dinners or wearing gloves.
I canceled my European training, too. Five days later, restrictions were imposed that would have forced them to be canceled, anyway. But at the time, my hosts were a bit annoyed with me.
I came home and isolated myself from my wife and mostly from the dog (dogs can become infected, too). Lenore is in lockdown. So is my son and his wife. I’m the only one who interacts with the public– doing all the shopping– and every time I do, I assume that I may have caught the virus. I change my clothes and take a shower when I get home. I live in my office. Lenore and I talk via Skype or else I stand on the stairs. She leaves food on the landing. No more movie nights on the couch with the family.
Do you think I’m overreacting? Wait two weeks before you say yes.
Remember that…?
In 1994 Intel shipped a bug with their Pentium chip. It would sometimes get the wrong answer when dividing numbers. It ultimately cost them $450 million. Before that happened, if a tester were to have said “Look at this bug, it might cost us millions!” he would have been told “where did you come up with that number? You are just overreacting.” After it happened, they invested millions in better testing. Intel was one of the first places I taught my testing class, in ’96.
Almost exactly ten years ago I was stuck in Europe for ten days because of a volcano in Iceland. Literally everyone flying in Europe had to cancel their flights and rebook (twice!). Before that happened, if a tester were to say “Let’s test what would happen if a million people suddenly rebook their flights” he would be ignored. I suspect testers got a temporary bump in influence, after that.
When bad things happen the world gets a little more open to the concerns of a tester; a little more open to the idea of testing at all.
COVID-19 on Orcas Island
According to the county, there is no coronavirus here. As I am writing this post, our first Covid-19 case on Orcas Island has been reported. Still, only 21 test results have come back, out of only 50 performed. Our county population is about 15,000 permanent residents. That’s poor test coverage. If we had comprehensive testing, and if all the people who tested positive were quarantined, my family would not need to lock down as hard as we have. Testing allows you to know the truth. Knowing the truth allows you to manage risk efficiently. No one can manage risk efficiently on Orcas Island, today, nor in most places around the world.
Testing lights the way, but we are blind. That leaves us with generalized fear. No wonder the stock markets are crashing!
Bugs in Your Product
Many companies spend little money or effort on testing. The people who push DevOps think they can automate all of it—they can’t and don’t. The people who push Agile says it’s everyone’s responsibility—which usually means no one takes responsibility to test well, to learn how to test well, to be ready to test well, and to coordinate testing well with everything else going on. When I challenge people on this, an answer I often get is “our way works… and we know that because we are still in business.”
As they discovered in Italy, and as we are all seeing right now, you can be “in business” just fine this week, be paralyzed next week, and be in deep mourning the week after that. Yes, you might be okay if your testing is bad, but you can’t know the quality of your testing just by how you have not yet suffered a terrible disaster.
I like the idea of automated output checks, and I like the idea of everyone helping to test. But beyond that, to be safe and responsible in our engineering, we need people who take testing responsibility seriously, and are given the time and resources to do it well.
Yet, ah! Why should they know their fate,
Since sorrow never comes too late,
And happiness too swiftly flies?
Thought would destroy their Paradise.
No more;—where ignorance is bliss,
‘Tis folly to be wise.
Thomas Gray was musing about innocence as he watched school boys playing in a yard. He was thinking, “Don’t break the spell. Let them be blissfully happy kids.” We, however, are not children. We should not wait until we are sure there is trouble. We must have faith in trouble; assume there is trouble until we can prove there isn’t. This helps us in our projects, in our testing, and it helps keep our families safe.
We are doing pretty much the same here in the Netherlands.
I don’t have it easy to isolate myself from the rest of my family, but luckily we are not on the risk group.
I told my parents to lock themselves in their countryside home with reserves and he went hesitantly… Then he called after the official lockdown, happy to be in a isolated area.
I bought a house and need to leave my current one next week… To move to the epicenter of the infection.
[James’ Reply: At least wear gloves. Also, I’m finding that whenever I am out in public, I’m keeping mental track of every single person who gets within 6 feet of me, and every single thing I touch or move. We will get good at this.]
Greetings James,
This entire situation is going to be the source of sooo much reflection, learning, root cause analysis, articles, studies, papers and on and on and on. It feels like a gigantic global anthropological study to me! Thanks for sharing some of your thoughts in this post.
You mentioned a couple situations where a tester raises a potential problem (Intel, Icelandic ash clouds) but is given short shrift. It makes me wonder, how much of a factor is timing and how should that impact the report of the potential problem?
It makes me think of the phrase “weak signal detection” or trying to identify the really important juicy information from background noise. When the folks in your Intel and Icelandic ash cloud examples raised the potential problem, how should they have done that depending on the different “noise” levels?
[James’ Reply: There are several ways to go. One way is to raise the volume. Get loud. But that is a rather crude and temporary solution. It’s also self-limiting, since people eventually tune you out. Another way is to tell a more VIVID story. Storytelling skill is paramount to influencing decision-makers. A good story isn’t successful because it’s louder, but rather because it feels plausible and relevant. The vividness of a product risk story must include its impact on people who matter. The degree to which the story reflects common biases and tropes also helps it seem plausible.]
It seems that as a context changes, peoples risk perception moves through phases of:
“that is so unlikely I don’t care.”
“definitely that’s possible but really unlikely so I don’t care.”
“this is a real problem so we’ll handle it!”
“why didn’t you tell me about this sooner!?”
“how can we avoid running into this situation in the future?”
Because those contexts are quite different, the way the information is shared in those moments probably impacts how it’s handled. To me it seems like a timing/advocacy/evidence/empathy problem as much as a “people don’t appreciate testing” problem.
Let me know what you think.
Cheers,
Vernon
(Side note: as I re-read before posting, the “phases” above seem to change focus. They start with a “probability” focus but seem to switch to “impact” by the end. That seems noteworthy to me.)
[James’ Reply: Yes, because the impact has happened and now it’s part of our collective tacit knowledge.]
Stay safe James.
If you think you are doing the right thing, then you are doing the right thing. Given what you already know, and what you already understand, your measures make total sense for me.
My wife is dealing with the spreading in her Hospital, as a Doctor in Preventive Medicine, she is dealing with outdated protocols, shortage of resources ( Masks, glasses .. ), but we still have conversations, what do we know, what can we try, how will we know if something works, what are the risks of trying or not trying…
As a tester, something I can do is read her protocols and apply my critical thinking, make questions, you know.
This spread is a bad one, and every day we don’t get ill, and we don’t make anyone get ill… is a good day.
Cheers and stay safe now, we’ll meet next time you come to Europe.
[James’ Reply: I bow to your wife. What she is about to face is a nightmare.]
Bill Gates on Covid-19:
“Testing is Everything”
May become my 2020 meme.
https://youtu.be/Xe8fIjxicoo
…and forgot the link.
What an urgent post James, and indeed very much linked to the telling the story about testing.
As it becomes clear now most governments have long been scaling down or in the case of my own government even sold their ability to test for and to develop vaccines for new viruses.
I hope these decisions will be investigated thoroughly by various media outlets that still play their role as guardians of our democracies.
As I’m quite safely working-from-home I think about all the workers that do not have this as an option.
Some because of the nature of their job, which we cannot value enough and often in terms of financial compensation we definitely do not value these jobs as a society.
But also all the people that have no choice but to go to work, because they have nothing to eat if they don’t. Sadly it’s these people all around the world, also in the more ‘developed parts’ that will suffer most from this lack of testing and the thoroughly stupid idea that ‘the free market’ is a cure for everything as if it were some law of physics.
I can only hope many people in the world come to realize that ‘the market’ will not do anything about this situation.
[James’ Reply: Thanks for the comment, man. Stay safe!]
I love the way you tie testing to risk. Testing serves risk management in several ways – helping to identify risks (by finding faults and limits) and helping to quantify risks (helping to inform how likely something is to occur).
Testing helps support informed risk management decisions.