“My job involves making highly business critical technical decisions,” says Alex Maclinovsky. “In my role, I have to go through a large number of technical documents, presentations and make a lot of time-sensitive decisions.”
Maclinovsky relies on a philosophy that has served him well throughout his career. He utilizes first principles derived from the sciences to make timely decisions in the world of applied technology.
Maclinovsky developed a grounding in the sciences during his college years. As a student at the Moscow Institute of Steel and Alloys, he studied under the theoretical physicist A. A. Abrikosov, who was awarded the Nobel Prize in 2003 for his theories on how matter behaves at extremely low temperatures.
A formative incident during the early days of home computing further shaped Maclinovsky’s belief in leaning on the pure sciences to make decisions related to applied technology. He recalls having lunch with a friend who had graduated from the biology department.
During the discussion, the two friends hit upon many similarities between computer and biological viruses. Both viruses were similar in size, and operated and evaded defenses in similar ways. Maclinovsky and his friend hypothesized how computer viruses could become even harder to defend against if they mimicked flu viruses: by rewriting themselves entirely, there was every danger that they would make it difficult, if not impossible, for antivirus programs to effectively combat them.
In this article, Maclinovsky outlines three scientific principles from fields as varied as physics and economics, and illustrates how engineers and technology can draw upon them to make important decisions.
1. The laws of thermodynamics: The laws of thermodynamics can allow you to understand why certain things are impossible without having to delve into the details. For example, related to the first law of thermodynamics, the law of conservation of energy states that the total energy of an isolated system remains constant. Furthermore, energy can neither be created or destroyed. Rather it can only be converted from one form to another.
The law of conservation of energy explains why it is impossible to create a perpetual motion machine – one that can do work indefinitely without an energy input. Over the years, there have been many ingenious designs proposed for the design of a perpetual motion machine. One approach proposed using the energy created by falling water onto a wheel to return the water back to the reservoir. However, such machines have proven to be practically infeasible because they violate the first law of thermodynamics.
Maclinovsky says that akin to the conservation of energy, every large-scale technological system is also governed by what he likes to call the “conservation of complexity.”
“Wayfair helps over 20 million customers create their unique feeling of home,” says Maclinovsky. “As you might imagine, the systems and technical architecture designed to serve the needs of such a large number of customers with highly individual tastes can be incredibly complex.”
Maclinovsky says that he reviews technical proposals that claim to be able to reduce complexity in one part of Wayfair’s system without having an impact on the remainder of Wayfair’s technical architecture.
“I don’t take these claims at face value because they would violate what I like to think of as a principle of the ‘conservation of complexity.’ In my experience, when you simplify one part of a system, it inevitably has an impact on other components by rendering them more complex. While this might not necessarily be a bad thing, it is important to have your eyes wide open in understanding the repercussions of your decision.”
2. The economics of internal organizations: Economists have historically focused on developing theories that explore how people interact with each other, and with organizations and institutions in society. However, in recent decades, they have also increasingly focused on how people and departments within an organization collaborate to achieve common goals. These goals can relate to making decisions related to the goods and services the firm should provide, the partners the company should collaborate with, and go-to-market strategies.
“Companies often set goals that are along the lines of ‘Let's make our app or site 10% faster than the leading competitor.’ These look like reasonable goals: everyone can agree that faster performance leads to increased customer satisfaction and ultimately increased revenue. The desire to be better than the competition is also part of human nature. But if one applies economic thinking, it becomes clear that decreasing latency beyond a certain point will not be noticeable by the customers, and will not result in an increase in revenue. However, each additional percent of decreased latency will be exponentially costlier than the last such increase. As a result, there is an optimal point beyond which improving performance will end up costing more than any resulting revenue gains.”
Maclinovsky is a champion of economic thinking among engineers at Wayfair. He gives the example of how economic thinking helped the firm control cloud computing costs effectively.
Like many firms that have experienced hypergrowth over the last decade, Wayfair has migrated its technology infrastructure from legacy systems to the cloud.
“Cloud computing has helped Wayfair transform and scale our storefront, and meet the needs of tens of millions of customers,” Maclinovsky says. “However, given the sheer size of our infrastructure, our leadership decided to implement measures that would help us cut our cloud-computing associated costs.”
Maclinovsky illustrates how economic thinking among the engineering team helped Wayfair avoid pitfalls faced in other organizations. For example, instead of resorting to cost-cutting across the company – a scenario that could lead to certain teams expending a lot of effort for minimal returns, the leadership team is experimenting with achieving efficiencies within the departments that have the greatest cloud-computing expenditures.
“Every engineer should strive to develop an understanding of the fundamentals of economics,” Maclinovsky says. “Developing a more holistic perspective is essential in your journey from an engineer to a technology leader.”
3. Taxonomy: A taxonomy is a hierarchical classification, in which things are organized into groups and categories. For example, The Tree of Life uses taxonomic categories to explore the evolution of life, and describe the relationships between living and extinct organisms. The classification surfaces a variety of insights that would otherwise be lost to the sheer number of organisms – such as organisms with vertebrae that have the same skeletal structure.
“Classifying concepts into categories is also an incredibly effective way of crafting compelling technical documents,” says Maclinovsky.
He points to a prior role in his career where he was responsible for helping organizations migrate to his company’s cloud computing solution. At the time, one of the world’s largest retail organizations had reservations about migrating to the new cloud solution, as there was no clearly defined documentation about how the system could help address the threat of DDoS attacks.
Maclinovsky conducted research on the most common types of DDoS attacks. He identified twenty-three kinds of DDoS attacks. To simplify matters, he placed them in four discrete groups based on commonalities in characteristics.
“I was able to clearly demonstrate that although we didn’t cover every possible type of attack, our company had ready-to-go solutions for all of the groups,” he says. “All we had to do was roll out a solution for a couple of additional kinds of DDoS attacks. The taxonomic categorization helped distill the information in a clear and understandable way to the retailer, and they migrated to our solution.”
“The value of an engineer is not merely in their technological knowledge. Rather, their value is proportional to how much they know outside their area of focus.”
These are just three examples of how engineers can leverage the sciences to make technology-related decisions. Maclinovsky recommends that engineers familiarize themselves with principles from a number of fields from astrophysics to psychology in their quest to become better engineers.
“The value of an engineer is not merely in their technological knowledge,” Maclinovsky says. “Rather, their value is proportional to how much they know outside their area of focus.”