Is the simplest approach always the best one?
JohnCrawfordIBM 100000BANX Visits (3432)
Is the simplest approach
by Timothy Burris
"Everything should be kept as simple as possible, but no simpler."
--attributed to Albert Eins
When I was teaching myself C++, I became fascinated with things like overloaded functions and smart pointers. Fortunately, an experienced programmer saw where I was heading, and exposed me to the power and elegance that simple solutions can offer.
The principle of parsimony, more commonly referred to as Occam’s Razor (after William of Ockham, c. 1287–1347), goes back at least as far as Aristotle (384-322 BCE), who stated:
"We may assume the superiority ceteris paribus [all things being equal] of the demonstration which derives from fewer postulates or hypotheses.”
This principle refers not to proven solutions, of course, but to “postulates or hypotheses.” Among competing hypotheses, the one that makes the fewest assumptions (i.e., that is the simplest) is the easiest to test for validity. This is not to say that the simplest explanation is always the right one.
Dr. Jason J. Braithwaite stressed the role of necessity over simplicity, as he described Occam’s razor in his paper “Occam’s Razor: The Principle of Parsimony” (2007). Dr. Braithwaite is associated with the Behavioural Brain Sciences Centre, School of Psychology, University of Birmingham, UK.)
“[Occam’s Razor is]… a logical tool that is employed to cut out irrelevant, unnecessary aspects from scientific arguments and philosophical systems. The central aspect of the principle is necessity. If it is not absolutely necessary to introduce certain complexities or hypothetical constructs into a given explanation, then don’t do it.”
In the day-to-day world, that kind of rigor isn’t always necessary. A simple explanation can be the best one, even if it’s based on questionable science. Consider a father who thinks he needs to explain to his seven-year-old daughter (he doesn’t) how to calculate the volume of water it will take to fill the swimming pool:
“Jennifer, there are three dimensions: length, width, and height. To calculate volume, you multiply those three dimensions by each other. Let’s measure the pool together and you can do the calculation, just like you do in school,” he said to her.
“Dad! Where have you been? There are, like, ten dimensions—at least! And you left out Time,” she responded.
Though dad’s simple explanation was not in line with current scientific thinking, it was sufficient to the task at hand, calculating the volume of water in the pool. Imagine trying to include all those other dimensions!
So what does this have to do with Information Technology?
Consider the case of an IT manager who is looking for the simplest solution that will provide her company everything it needs in the way of application monitoring. She is discussing a trial of the IBM SaaS Application Performance Management offering with her lead operator:
“John, simpler is better. This SaaS offering lets me see application health at a glance.
"This would be great for you, too. There’s almost nothing to install. And they have agents for all the components in our applications, with more being added all the time. Wouldn’t that give you what you need to do your job?,” the manager said.
“I agree it’s impressive, and it would enable me to do part of my job. But our team also needs the on-premises SCAPM infrastructure to provide you with things we can’t get from SaaS, at least not yet. You know how valuable that trending analysis has been for us, helping us spot potential problems while all systems are still green. That requires the on-prem infrastructure. The same is true of the integration we do with some legacy monitoring systems. And there are more,” the operator responded.
The hybrid solution advocated by the operator isn’t the simplest solution available, but it is the solution best suited to this company’s needs.
I’ll conclude with an observation of the English mathematician and philosopher, Alfred North Whitehead from his writing in The Concept of Nature (1926):
“The aim of science is to seek the simplest explanations of complex facts. We are apt to fall into the error of thinking that the facts are simple because simplicity is the goal of our quest. The guiding motto in the life of every natural philosopher should be, ‘Seek simplicity and distrust it.’”
I would argue that the same is true in the world of systems management.
Timothy Burris is a technical enablement specialist with IBM Cloud & Smarter Infrastructure. He came to IBM from the IntelliWatch R&D group, as part of the Candle acquisition. He joined Technical Enablement via TechWorks and Tech Sales.