Baby Steps in Science
Our baby taught me about the scientific method.
“You can’t just put your ear plugs in!”, exclaimed my wife.
“How else do I sleep with the constant crying?”, I rebutted, a mild sense of guilt riding in on horseback.
“Have you considered that his crying is merely a consequence of something else?”
“Urgh. Stop being so reasonable. So, how do we work out why he’s crying?”, I croaked.
“We need to be scientists. We first guess the cause of the crying. Then, we ask ourselves: what else - in addition to crying - would we expect as a consequence of the speculated cause? Finally, we check to see if those consequences exist. If so, our speculated cause was correct! Can you smell something funky?”
As I sniffed the air and realised what needed to be done, something in my head clicked.
When Richard Feynman worked at identifying scientific laws to explain natural phenomena, he followed three steps:
Suggests Feynman, “First, we guess the law. Don’t laugh - that’s really true. Then we compute the consequences of the guess, to see what this law would imply. [...] And then we compare the computation results to experiment [reality] to see if the law works.
If the law disagrees with experiment, it's wrong. And that simple statement is the key to science.
It doesn't make a difference how beautiful your guess is. It doesn't make a difference how smart you are [...]. If the law disagrees with experiment, it's wrong.”
In organisations it is all too easy to try to tackle the issues we face head on, and forget they’re usually a consequence of something else. We may tackle increasing IT costs with outsourcing, or an increasing number of IT outages with more stringent software testing. At best, we waste effort. More likely, we make things worse.
The takeaway? Increasing IT costs and outages - like crying babies and bad smells - do not happen without reason, and are instead consequences of something we need to understand. Failing to do so is as effective as wearing nose and ear plugs when your baby’s nappy needs changing.
